kevchess15/deepseek-harness
1
1
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions 1
deepseek-harness/docs/module-graph.md

196 lines
6 KiB
Markdown
Raw Normal View History

feat: generate module dependency graph with freshness gate Add scripts/gen-module-graph.ts, which derives the inter-package dependency graph from each package's @deepseek-ai/dsh-* peerDependencies and renders docs/module-graph.md (a GitHub-native Mermaid graph plus a dependency table). Output is deterministic so a regenerate-and-diff check is stable. Wire a freshness gate the same way doc-sync is wired (ADR 0007: hooks and CI run the same package.json scripts): verify-module-graph runs in pre-push (lefthook) and as a CI step. It fails if the committed file drifts from what the generator would produce.
2026-06-16 21:00:24 +08:00
<!-- Generated by scripts/gen-module-graph.ts — do not edit by hand.
Run `pnpm run gen-module-graph` to regenerate. -->
# Module dependency graph
docs: document module-graph gate; unwrap generated prose Codex review follow-up: - Update docs/development.md: add verify-module-graph to the pre-push and CI gate lists and gen/verify-module-graph to the daily commands, so the contributor guide matches the new freshness gate. - Emit the module-graph.md intro paragraph as a single line (repo Markdown convention: one line per paragraph), since regenerating would otherwise reintroduce hard wrapping.
2026-06-16 21:21:21 +08:00
Inter-package dependencies among the `@deepseek-ai/dsh-*` harness packages, derived from each package's `peerDependencies` (the canonical runtime-dependency signal). An edge `a --> b` means package `a` depends on package `b`. Names have the `@deepseek-ai/dsh-` prefix stripped.
feat: generate module dependency graph with freshness gate Add scripts/gen-module-graph.ts, which derives the inter-package dependency graph from each package's @deepseek-ai/dsh-* peerDependencies and renders docs/module-graph.md (a GitHub-native Mermaid graph plus a dependency table). Output is deterministic so a regenerate-and-diff check is stable. Wire a freshness gate the same way doc-sync is wired (ADR 0007: hooks and CI run the same package.json scripts): verify-module-graph runs in pre-push (lefthook) and as a CI step. It fails if the committed file drifts from what the generator would produce.
2026-06-16 21:00:24 +08:00
```mermaid
graph TD
feat(types): brand bash ids + stop brand erosion; extract Branded to dsh-brand Type-only change (brands are zero-cost casts; no runtime/wire impact). Closes the two gaps in the "brand ids that cross package boundaries" policy and fixes the dependency direction so a capability package never pulls in an unrelated one. - Extract the `Branded<B>` primitive into a new standalone type-only package `@deepseek-ai/dsh-brand` (packages/util/brand) with no harness-package deps. dsh-llm keeps its owned CallId but imports Branded from dsh-brand; dsh-session, dsh-agent, and dsh-bash all import Branded from there. dsh-bash depends on dsh-brand ALONE — never on dsh-llm or dsh-session (the architectural fix: a generic execution backend must not couple to the LLM or session vocabulary). - Mint BashTaskId + OwnerToken in dsh-bash and thread them through BashTask.id, the get/ownerOf/list/readOutput/kill seam, the bash-local generation site, and the dsh-tool-bash validate/access surface. OwnerToken is a DISTINCT brand from SessionId so the seam stays decoupled; dsh-tool-bash is the single boundary that casts SessionId -> OwnerToken. - Brand at the SOURCE, not via mid-pipeline casts: agent-loop's Config types agents[].id as AgentId and resumeSessionId as SessionId, so the brand enters at the config boundary and the inner create()/resume casts disappear (only the genuinely-new per-run session-id string is cast). - Stop brand erosion: propagate CallId/SessionId/AgentId to the registry/store Map keys and public params/exports (SessionStore, AgentRegistry + factory options, the ACP session-id surface + ToolPresenter CallId map, the persistence coordinator, invariants pendingCalls, the pi-ai tool-call maps). - Docs: document BashTaskId/OwnerToken in bash.md (type-equiv re-pasted), point the Branded type-equiv at dsh-brand, fix stale param types in the session/ agent/bash READMEs, regenerate the cordis catalog + module graph. Implements docs/rfc/proposed/architecture/2026-06-20-branded-ids.md
2026-06-21 07:17:25 +08:00
bash --> brand
llm --> brand
refactor: order module-graph table topologically (low→high level) Replace the alphabetical package ordering with a Kahn-style topological sort (alphabetical tiebreak within each layer to stay deterministic), so the table and graph list leaf interfaces first and dependents last.
2026-06-16 21:11:09 +08:00
bash-local --> bash
refactor(fs): split filesystem seam into provider ctx.fs + policy ctx.fileContext Implements the split-the-filesystem-seam RFC. ctx.fs shrinks to a text-storage provider seam (resolve/stat/readText/streamText/writeText/editText with branded FsTargetKey/FsVersion and an explicit FsWriteExpectation); the new dsh-file-context package owns the model-facing policy (read windowing, observed-state, write/edit freshness) as the concrete ctx.fileContext service. Authorization is now freshness-based rather than full/partial view: a windowed read records the file version and authorizes a later edit when the file is unchanged, removing the dead-end where reading lines 100-150 of a large file could not edit line 120. editText stays a provider primitive so version guard + literal match + atomic rewrite remain one critical section, and the stale check runs before matching so a stale edit reports FS_STALE_VERSION. tool-fs injects fileContext, never reaching around to ctx.fs (the no-bypass contract).
2026-06-26 17:23:18 +08:00
fs --> brand
Add filesystem capability seam and tools
2026-06-22 10:48:41 +08:00
fs --> llm
refactor: order module-graph table topologically (low→high level) Replace the alphabetical package ordering with a Kahn-style topological sort (alphabetical tiebreak within each layer to stay deterministic), so the table and graph list leaf interfaces first and dependents last.
2026-06-16 21:11:09 +08:00
llm-deepseek --> llm
llm-pi-ai --> llm
feat(types): brand bash ids + stop brand erosion; extract Branded to dsh-brand Type-only change (brands are zero-cost casts; no runtime/wire impact). Closes the two gaps in the "brand ids that cross package boundaries" policy and fixes the dependency direction so a capability package never pulls in an unrelated one. - Extract the `Branded<B>` primitive into a new standalone type-only package `@deepseek-ai/dsh-brand` (packages/util/brand) with no harness-package deps. dsh-llm keeps its owned CallId but imports Branded from dsh-brand; dsh-session, dsh-agent, and dsh-bash all import Branded from there. dsh-bash depends on dsh-brand ALONE — never on dsh-llm or dsh-session (the architectural fix: a generic execution backend must not couple to the LLM or session vocabulary). - Mint BashTaskId + OwnerToken in dsh-bash and thread them through BashTask.id, the get/ownerOf/list/readOutput/kill seam, the bash-local generation site, and the dsh-tool-bash validate/access surface. OwnerToken is a DISTINCT brand from SessionId so the seam stays decoupled; dsh-tool-bash is the single boundary that casts SessionId -> OwnerToken. - Brand at the SOURCE, not via mid-pipeline casts: agent-loop's Config types agents[].id as AgentId and resumeSessionId as SessionId, so the brand enters at the config boundary and the inner create()/resume casts disappear (only the genuinely-new per-run session-id string is cast). - Stop brand erosion: propagate CallId/SessionId/AgentId to the registry/store Map keys and public params/exports (SessionStore, AgentRegistry + factory options, the ACP session-id surface + ToolPresenter CallId map, the persistence coordinator, invariants pendingCalls, the pi-ai tool-call maps). - Docs: document BashTaskId/OwnerToken in bash.md (type-equiv re-pasted), point the Branded type-equiv at dsh-brand, fix stale param types in the session/ agent/bash READMEs, regenerate the cordis catalog + module graph. Implements docs/rfc/proposed/architecture/2026-06-20-branded-ids.md
2026-06-21 07:17:25 +08:00
session --> brand
refactor: order module-graph table topologically (low→high level) Replace the alphabetical package ordering with a Kahn-style topological sort (alphabetical tiebreak within each layer to stay deterministic), so the table and graph list leaf interfaces first and dependents last.
2026-06-16 21:11:09 +08:00
session --> llm
system-prompt --> llm
chore: adapt web seam to master's tsconfig + regenerate catalogs Register the five web packages in the root tsconfig.json project graph (master's typecheck moved to `tsc -b tsconfig.json` and dropped the separate tsconfig.typecheck.json), and regenerate the module graph and cordis catalog so they reflect the web packages on master.
2026-06-26 19:18:46 +08:00
web --> llm
feat(types): brand bash ids + stop brand erosion; extract Branded to dsh-brand Type-only change (brands are zero-cost casts; no runtime/wire impact). Closes the two gaps in the "brand ids that cross package boundaries" policy and fixes the dependency direction so a capability package never pulls in an unrelated one. - Extract the `Branded<B>` primitive into a new standalone type-only package `@deepseek-ai/dsh-brand` (packages/util/brand) with no harness-package deps. dsh-llm keeps its owned CallId but imports Branded from dsh-brand; dsh-session, dsh-agent, and dsh-bash all import Branded from there. dsh-bash depends on dsh-brand ALONE — never on dsh-llm or dsh-session (the architectural fix: a generic execution backend must not couple to the LLM or session vocabulary). - Mint BashTaskId + OwnerToken in dsh-bash and thread them through BashTask.id, the get/ownerOf/list/readOutput/kill seam, the bash-local generation site, and the dsh-tool-bash validate/access surface. OwnerToken is a DISTINCT brand from SessionId so the seam stays decoupled; dsh-tool-bash is the single boundary that casts SessionId -> OwnerToken. - Brand at the SOURCE, not via mid-pipeline casts: agent-loop's Config types agents[].id as AgentId and resumeSessionId as SessionId, so the brand enters at the config boundary and the inner create()/resume casts disappear (only the genuinely-new per-run session-id string is cast). - Stop brand erosion: propagate CallId/SessionId/AgentId to the registry/store Map keys and public params/exports (SessionStore, AgentRegistry + factory options, the ACP session-id surface + ToolPresenter CallId map, the persistence coordinator, invariants pendingCalls, the pi-ai tool-call maps). - Docs: document BashTaskId/OwnerToken in bash.md (type-equiv re-pasted), point the Branded type-equiv at dsh-brand, fix stale param types in the session/ agent/bash READMEs, regenerate the cordis catalog + module graph. Implements docs/rfc/proposed/architecture/2026-06-20-branded-ids.md
2026-06-21 07:17:25 +08:00
agent --> brand
feat: generate module dependency graph with freshness gate Add scripts/gen-module-graph.ts, which derives the inter-package dependency graph from each package's @deepseek-ai/dsh-* peerDependencies and renders docs/module-graph.md (a GitHub-native Mermaid graph plus a dependency table). Output is deterministic so a regenerate-and-diff check is stable. Wire a freshness gate the same way doc-sync is wired (ADR 0007: hooks and CI run the same package.json scripts): verify-module-graph runs in pre-push (lefthook) and as a CI step. It fails if the committed file drifts from what the generator would produce.
2026-06-16 21:00:24 +08:00
agent --> llm
agent --> session
feat(compact): compaction capability seam — abstract CompactService interface Adds the @deepseek-ai/dsh-compact interface package: the abstract CompactService (ctx.compact) with compactIfNeeded / compactRegion, the compact/* session-event types via SessionEventMap declaration merging, and the capability-seam RFC. Wires the package into the three root tsconfigs and the cordis catalog. A backend implementation lands separately.
2026-06-22 14:52:30 +08:00
compact --> llm
compact --> session
Add filesystem capability seam and tools
2026-06-22 10:48:41 +08:00
fs-local --> fs
fix(fs): address review — rename to dsh-fs-policy, fs/*-intent events, RFC currency, ENOTDIR Rename per review naming decisions: - package dsh-file-context → dsh-fs-policy (dir, package name, plugin name, tsconfig refs, importers, type-equiv manifest, generated catalog + module-graph) - events fs/write-expectation → fs/write-intent, fs/edit-expectation → fs/edit-intent (fs/observed unchanged); type FsWriteExpectation → FsWriteIntent, "expectation" wording → "intent" throughout - exported FileContextExec → FsPolicyExec Make the implemented RFCs describe what shipped, not the superseded designs: the 2026-06-17 capability-seam + tool-schemas RFCs no longer place policy on ctx.fs or use full/partial-view authorization, and the fsspec RFC's ctx.fileContext service prose is rewritten to the fs/* event-gate reality (freshness-based auth). Sharpen docs/rfc/implemented/AGENTS.md: a rename is a fact to fix IN PLACE — the "new RFC" escape hatch is for macro decision reversals only, not renames. Code fixes from review: - fsio.ts resolveLocalTarget/probe translate ENOTDIR (a parent path segment is a file) into the structured FsError taxonomy instead of leaking a raw Node error; resolve reports FS_NOT_FOUND, probe reports absent. Regression tests proven to fail on the unfixed code. - tool-fs HMR test now asserts prompt sections (not just tool schemas) are withdrawn on disposal. - fs/observed is a plain (unguarded) ctx.emit: correct the fs-policy comment, filesystem.md, and tool-fs module doc that wrongly claimed the tool "contains" a throwing listener; a throw surfaces as the tool's isError result. - drop the false "loaded by the default product config" claim (no config wires the fs tools yet), the duplicate ctx.bash service-map row, the stale FileReadRequest catalog link-map entry, and the fs/fs README EOF blank line; correct the dsh-fs package.json description.
2026-07-02 03:12:38 +08:00
fs-policy --> fs
feat(hooks): dsh-hook-protocol — shared Claude Code / Codex hook wire-protocol core The two hook bridges (dsh-hooks-claude, dsh-hooks-codex) would otherwise duplicate the bulk of the protocol — Codex deliberately reimplements a SUBSET of the Claude Code protocol (same hooks.json shape, exit-code/stdout contract, command-hook model). This library holds the genuinely-identical primitives; each bridge owns only what differs (per-event stdin payload, env/substitution, decision mapping). New packages/hooks/ group; hook-protocol is a LIBRARY (no plugin, registers/injects nothing): - matcher: matchesMatcher(pattern, query, mode) — the one dialect axis collapsed to a mode param (claude = literal-or-regex with pipe alternation; codex = always unanchored regex). Match-all on absent/''/'*'; invalid regex matches nothing. - codec: parseHookOutput(exit, stdout, stderr) → dialect-neutral HookOutput. Exit 0 → lenient JSON; exit 2 → blocking error (stderr = reason, surfaced as decision:'block'); other → non-blocking. Parses the CC superset (continue/stopReason/decision/hookSpecificOutput.{permissionDecision, additionalContext,updatedInput}/systemMessage); permissionDecision overrides the legacy top-level decision. - runner: runHook(bash, hook, opts, now) — runs a command hook via ctx.bash (stdin payload + trusted-plugin env), honors timeoutSec, never throws (executor reject → non-blocking-error HookOutput). Injected clock for testable durations. - merge: mergeHookOutputs — most-restrictive fold (deny>ask>allow, sticky stop, block reasons joined, context/system-messages accumulated). - hook/* session events (declaration-merged into SessionEventMap, log-only like compact/*) + appendHookInvoked/appendHookResult helpers. updatedInput is parsed but NOT honored (deferred pre-tool-input-rewrite RFC); a bridge logs+warns. 47 unit tests at per-file 100% (matcher per-mode, codec per exit-code/field, runner plumbing w/ stub executor, merge precedence, hook/* helpers). RFC: implemented/feature/2026-06-30-hook-protocol-lib.md.
2026-07-01 00:38:06 +08:00
hook-protocol --> bash
hook-protocol --> session
refactor(examples): extract reusable logic into tested packages Logic that lived under examples/ was outside the per-file 100% coverage gate (examples/ are not workspaces) and, in the stdio-UI case, duplicated across two examples. Move it into packages/ so it is gated and de-duped. - packages/ui-stdio (new): unify the two diverged stdio-chat.ts copies into one @deepseek-ai/dsh-ui-stdio plugin (welcome/agent Config). A test-only I/O seam (createStdioChat(ctx, config, runtime)) keeps process streams out of the serializable config and makes every render/EOF/disposal branch unit-testable. Per-file 100%. echo/coding cordis.yml now load the package; both src/stdio-chat.ts deleted. - packages/llm-replay (new): move examples/acp-agent/src/llm-replay.ts (+ its spec) here so its derive/parse/replay branches fall under the coverage gate. cordis.snapshot.yml + README rewired to the package name; added apply/env /assertNever/abort tests to reach per-file 100%. - examples/{echo,coding}-agent: keyless Loader-path e2e smokes that boot the real cordis.yml (no key) — the guard a hand-mounted unit test cannot be for the unwrapExports/export-shape class (postmortem 0001). examples/AGENTS.md codifies the keyless+with-key smoke convention (keyless-by-nature exception for echo-agent). - AGENTS.md: a scoped, removal-triggered pre-release stance (foundation over blast radius). packages/README.md: new rows + a FIXME to later regroup ALL packages into a hierarchy. Wiring: tsconfig paths/refs, publint, knip, module-graph. Verified: typecheck, lint, test:coverage (887 tests, 100%), build, hygiene, doc-sync, test:snapshot (10), test:e2e (6 keyless pass, with-key self-skip).
2026-06-19 12:42:28 +08:00
llm-replay --> llm
llm-replay --> session
Merge remote-tracking branch 'origin/master' into split/session-persistence
2026-06-16 23:16:59 +08:00
session-persistence --> session
chore: adapt web seam to master's tsconfig + regenerate catalogs Register the five web packages in the root tsconfig.json project graph (master's typecheck moved to `tsc -b tsconfig.json` and dropped the separate tsconfig.typecheck.json), and regenerate the module graph and cordis catalog so they reflect the web packages on master.
2026-06-26 19:18:46 +08:00
web-fetch-local --> web
feat(web): add DeepSeek-backed web search provider Add @deepseek-ai/dsh-web-search-deepseek: a WebSearchProvider that calls DeepSeek's Anthropic-compatible Messages API with the native web_search_20250305 server tool and parses the structured web_search_tool_result blocks into the ctx.web seam's WebSearchResult. - Namespace plugin (inject: ['web']), no default export — registers into ctx.web like dsh-llm-deepseek registers into ctx.llm. - Strict mode: a response with no web_search_tool_result block throws WEB_PROVIDER_ERROR rather than scraping URLs from model prose. - Reuses $DEEPSEEK_API_KEY; baseURL defaults to the Anthropic-compatible base (api.deepseek.com/anthropic/v1) and does NOT reuse $DEEPSEEK_BASE_URL, which belongs to the chat-completions LLM adapter. - snippet joined from text-block citations; sources deduped by url. - Two-stage build layout (outDir lib/types) matching the other web packages; registered in tsconfig.json, tsconfig.build.json, knip.json, and docs/module-graph.md.
2026-06-29 15:21:12 +08:00
web-search-deepseek --> web
chore: adapt web seam to master's tsconfig + regenerate catalogs Register the five web packages in the root tsconfig.json project graph (master's typecheck moved to `tsc -b tsconfig.json` and dropped the separate tsconfig.typecheck.json), and regenerate the module graph and cordis catalog so they reflect the web packages on master.
2026-06-26 19:18:46 +08:00
web-search-exa --> web
web-search-perplexity --> web
feat(compact-basic): baseline compaction backend (squashed from compact-basic) Collapses the per-round review churn of the prior compact-basic branch into a single clean baseline on top of compact-interface, so the upcoming retention refactor lands as fresh, well-scoped commits rather than stacking on a history of fixes that are being superseded.
2026-06-25 17:30:42 +08:00
compact-basic --> agent
compact-basic --> compact
compact-basic --> llm
compact-basic --> session
refactor: order module-graph table topologically (low→high level) Replace the alphabetical package ordering with a Kahn-style topological sort (alphabetical tiebreak within each layer to stay deterministic), so the table and graph list leaf interfaces first and dependents last.
2026-06-16 21:11:09 +08:00
invariants --> agent
invariants --> llm
invariants --> session
Merge remote-tracking branch 'origin/master' into split/session-persistence
2026-06-16 23:16:59 +08:00
session-persistence-jsonl --> session
session-persistence-jsonl --> session-persistence
Merge remote-tracking branch 'origin/split/agent-factory' into split/session-persistence-sqlite
2026-06-16 23:19:45 +08:00
session-persistence-sqlite --> session
session-persistence-sqlite --> session-persistence
refactor: order module-graph table topologically (low→high level) Replace the alphabetical package ordering with a Kahn-style topological sort (alphabetical tiebreak within each layer to stay deterministic), so the table and graph list leaf interfaces first and dependents last.
2026-06-16 21:11:09 +08:00
tools --> agent
tools --> llm
tools --> system-prompt
workflow: dynamic workflows — script-driven multi-agent orchestration A new capability family at packages/workflow/ in the bash seam shape, modeled on Claude Code's dynamic workflows: the model writes a JavaScript orchestration script (export const meta = {...} + plain-JS body), a runtime executes it, and the script — not the conversation — holds the loop, the branching, and the intermediate results. - dsh-workflow (ctx.workflows): abstract WorkflowService + run vocabulary (WorkflowRun whose result NEVER rejects) + observe-only workflow/* events carrying data snapshots (id + meta, never the live run), per-listener contained like subagent/*. - dsh-workflow-vm: in-process node:vm engine. Meta extraction via a string/comment-aware scanner (template interpolation rejected; literal evaluated alone in an empty timed context; statement blanked line- preservingly so stacks keep script line numbers). Hooks: agent(prompt, {label, phase, schema, model}) over ctx.subagents, parallel(), pipeline() (no cross-stage barrier), phase(), log(), args. Fatal-vs-null discipline: hook misuse (unknown/deferred options, bad arguments, unsupported schemas, tripped caps, seam start failures, cancellation) throws fatal WorkflowErrors the combinators RE-THROW — never dissolved into the per-item null reserved for child failures. Realm boundary: inbound values materialized by descriptor walks that never invoke accessors (defineProperty copies, __proto__-safe); outbound values rebuilt in-realm via the context's own JSON.parse. Determinism bans (Date.now/Math.random/argless new Date) kept so future resume support cannot break scripts. Caps and timeouts are validated Config. Every hook promise carries a no-op rejection consumer (app-boot exits on unhandled rejections). - dsh-tool-workflow: the model-facing workflow tool, synchronous like dsh-tool-subagent (start → await → try/finally dispose; abort bridged; non-completed → isError). Generic render card titled by a textual meta.name sniff. The tool description carries the authoring contract. Wired into examples/{coding-agent,acp-agent} with explicit-ask-only guidance. Coverage at every tier: unit (meta scanner, materializer incl. counting-getter and __proto__ regressions, combinator semantics, concurrency ceiling, caps, cancellation, no-unhandled-rejection abandon), integration over the real spawn stack, with-key e2e (real two-phase run + the tool through the registry pipeline), and a recorded ACP snapshot scenario (workflow-run, 1 child session). RFC: docs/rfc/implemented/feature/2026-07-05-dynamic-workflows.md (deferred work explicitly listed). AGENTS.md budget 1575 → 1590 for the new group's layout line.
2026-07-05 13:29:35 +08:00
workflow --> agent
workflow --> brand
workflow --> llm
feat(acp): tool-owned tool-call UI presentation (title/command/output) In Zed the tool-call card showed only "bash" — the bare tool name — instead of what the command does. Fix it by letting each TOOL own how its calls render, rather than the bridge special-casing names. dsh-tools: add an optional two-state presentation seam to ToolDefinition / defineTool — `presentCall(args)` (pending: title, kind, rawInput) and `presentResult(args, result)` (completed: title?, content?). Provider-neutral `ToolCallKind`/`ToolCallPresentation`/`ToolResultPresentation` vocabulary so tools never depend on ACP. defineTool soft-validates args (display runs on log replay, so a malformed/old shape returns undefined instead of throwing). dsh-tool-bash: bash declares presentCall (model `description` → title, exact `command` → rawInput, kind execute) and presentResult (wrap output in a fenced ```console block — a UI-only affordance kept out of the model-facing result); bash_output/bash_kill present task-scoped titles. dsh-acp: inject `tools`; a per-session `ToolPresenter` looks the tool up by name and maps its neutral presentation to the ACP tool_call/tool_call_update wire shape, with a generic fallback (title = name) for tools that declare nothing. Because the `tool/result` event carries only {callId, content, isError}, the presenter keeps a small bridge-local map of ONLY in-flight calls' (name, args), keyed by callId and removed as each result is presented — no event-schema or core change. Replay uses a throwaway presenter so loaded sessions render identically to live ones. Tests: dsh-tools defineTool presenters (typed args, soft-validate), tool-bash bash/bash_output/bash_kill presenters, acp ToolPresenter (tool-owned mapping, unknown-callId fallback, in-flight-only map), and an end-to-end turn through the bridge. The key-gated e2e now asserts a real bash call's title is the model description (not "bash") and rawInput is the command — verified against the real DeepSeek model. The test harness derives its inject from the bridge's exported `inject` so it can't drift again.
2026-06-18 09:01:36 +08:00
acp --> agent
acp --> llm
acp --> session
acp --> session-persistence
acp --> tools
feat: generate module dependency graph with freshness gate Add scripts/gen-module-graph.ts, which derives the inter-package dependency graph from each package's @deepseek-ai/dsh-* peerDependencies and renders docs/module-graph.md (a GitHub-native Mermaid graph plus a dependency table). Output is deterministic so a regenerate-and-diff check is stable. Wire a freshness gate the same way doc-sync is wired (ADR 0007: hooks and CI run the same package.json scripts): verify-module-graph runs in pre-push (lefthook) and as a CI step. It fails if the committed file drifts from what the generator would produce.
2026-06-16 21:00:24 +08:00
agent-loop --> agent
agent-loop --> llm
agent-loop --> session
Merge remote-tracking branch 'origin/split/session-persistence' into split/agent-factory
2026-06-16 23:18:31 +08:00
agent-loop --> session-persistence
feat: generate module dependency graph with freshness gate Add scripts/gen-module-graph.ts, which derives the inter-package dependency graph from each package's @deepseek-ai/dsh-* peerDependencies and renders docs/module-graph.md (a GitHub-native Mermaid graph plus a dependency table). Output is deterministic so a regenerate-and-diff check is stable. Wire a freshness gate the same way doc-sync is wired (ADR 0007: hooks and CI run the same package.json scripts): verify-module-graph runs in pre-push (lefthook) and as a CI step. It fails if the committed file drifts from what the generator would produce.
2026-06-16 21:00:24 +08:00
agent-loop --> system-prompt
agent-loop --> tools
feat(hooks): dsh-hooks-claude + dsh-hooks-codex bridges (hooks stack PR-F) The two bridge plugins that run a user's existing Claude Code / Codex hook config on the harness's typed interception seams, built on the shared dsh-hook-protocol library. A bridge is a faithfulness adapter, not a power tool: anything it does a native cordis plugin does more powerfully — the bridge exists only to run UNMODIFIED external hooks. - dsh-hooks-claude: CC dialect. Seven hook points (SessionStart, UserPromptSubmit, PreToolUse, PostToolUse, Stop, SubagentStart, SubagentStop), CC per-event stdin payloads, env + ${CLAUDE_PLUGIN_ROOT}/ ${CLAUDE_PROJECT_DIR} substitution, literal-or-regex matcher. - dsh-hooks-codex: Codex dialect — a deliberate subset. Five hook points, always-regex matcher, snake_case payloads (turn_id/model, no trailing newline), no env/substitution, block-only decisions. Both map the neutral merged outcome onto the seam's typed Decision and stamp an explicit {kind:'plugin'} source on injected context (so it is never mislabeled as a user prompt). Config parse-failure is contained; only command hooks run. updatedInput is logged+warned (input rewrite deferred); the Stop loop-guard is deferred (TODO). Tests: per-file 100% — config-parse unit branches + per-seam mappings end-to-end through the REAL loop + REAL bash + REAL shell scripts (scripted mock model only) + a real-Loader export-shape guard. A keyless ACP snapshot scenario (hook-prompt-block) proves a UserPromptSubmit hook blocks a prompt end-to-end (rejected turn -> ACP cancelled, hook/* events in the log); a with-key e2e (hooks.e2e.ts) proves a PreToolUse hook blocks real bash (verified on disk). The snapshot normalizer now scrubs hook/result.durationMs. RFC: docs/rfc/implemented/feature/2026-06-30-hook-bridges.md
2026-07-01 04:22:00 +08:00
hooks-codex --> agent
hooks-codex --> hook-protocol
hooks-codex --> llm
hooks-codex --> session
hooks-codex --> tools
Add subagent capability seam: interface, mock backend, model-facing tool Introduce the `packages/subagent/` group and the abstract subagent seam — an agent delegating to a child agent — as a named-provider registry (`ctx.subagents`), unlike the single-implementation bash seam, so multiple transports (in-process, ACP, future A2A) coexist. This first PR lands the interface, a scripted test backend, and the model-facing tool, validated through the real cordis load path. - dsh-subagent: SubagentService registry + SubagentProvider/SubagentRun vocabulary + subagent/start|end events. Start-time capabilities (outputSchema, depthLimit, toolFilter) are checked pre-start and rejected loud; runtime capabilities (sendMessage, resume) are optional methods on SubagentRun. - dsh-subagent-mock (support): scripted provider for keyless, deterministic tests through the real Loader/export path. - dsh-tool-subagent: the model-facing `subagent` tool, config-bound to one provider; synchronous collect with try/finally dispose, signal->cancel bridging, and non-completed-stop-reason -> isError mapping. - Proposed RFC documenting the seam, the fork-vs-spawn-as-separate-backends decision, own-session isolation, synchronous-collect scope, and the deferral of background/poll/spill to a future unification with bash. - Wire the new group into tsconfigs, build refs, package hierarchy docs, the module graph, and the cordis catalog. RFC: docs/rfc/proposed/feature/2026-06-21-subagent-capability-seam.md
2026-06-21 22:31:56 +08:00
subagent --> agent
subagent --> llm
subagent --> tools
feat: generate module dependency graph with freshness gate Add scripts/gen-module-graph.ts, which derives the inter-package dependency graph from each package's @deepseek-ai/dsh-* peerDependencies and renders docs/module-graph.md (a GitHub-native Mermaid graph plus a dependency table). Output is deterministic so a regenerate-and-diff check is stable. Wire a freshness gate the same way doc-sync is wired (ADR 0007: hooks and CI run the same package.json scripts): verify-module-graph runs in pre-push (lefthook) and as a CI step. It fails if the committed file drifts from what the generator would produce.
2026-06-16 21:00:24 +08:00
tool-bash --> agent
tool-bash --> bash
tool-bash --> llm
tool-bash --> tools
Add filesystem capability seam and tools
2026-06-22 10:48:41 +08:00
tool-fs --> fs
tool-fs --> llm
feat(tool-fs): result-time applied-hunk diffs for write/edit fs write/edit now emit a result-time contextual-diff tool_call_update (the applied hunk with ±3 context lines, one hunk per replace_all site), matching what claude-agent-acp sends and what makes an editor render the change in place. The call-time snippet diff stays; the result hunk supersedes it (ACP content-replace). Mechanism: - A persisted tool-private `meta` channel: execute may return `{ content, meta }`; `meta` (JsonValue) rides on the tool/result event and is handed back to presentResult, so the diff reproduces on replay (event-sourced). JsonValue is now exported from dsh-session. - The backend returns raw before/after text (storage facts) on FsWriteOutcome/FsEditOutcome; the tool computes the hunk via the npm `diff` package's structuredPatch. A create has no before → no result diff; a failed/aborted mutation carries no meta. - ToolResultView gains a DiffResultView; the bridge's result-side switch renders it as {type:'diff'} content blocks. RFC: docs/rfc/implemented/architecture/2026-07-02-result-time-applied-hunk-diffs.md (justifies the npm `diff` runtime dep over vendoring and the meta channel); the render-intent-union RFC's Non-goal is updated to record this shipped. All fs snapshot goldens re-recorded; edit/overwrite gain the contextual result diff, create/read/policy-reject unchanged in structure.
2026-07-03 17:12:00 +08:00
tool-fs --> session
Add filesystem capability seam and tools
2026-06-22 10:48:41 +08:00
tool-fs --> system-prompt
tool-fs --> tools
feat(tool-todo): add the model-facing todo_write tool Add @deepseek-ai/dsh-tool-todo (a new packages/todo/ group): a model-facing todo_write(todos: [{content, status}]) tool with whole-list-replace semantics. Each call appends the full list as a todo/write event to the calling agent's session log; the current list is the most recent such event (last-write-wins). Single-owner — a non-agent caller is rejected. Beyond the schema's type/required/enum checks, execute rejects empty/duplicate content and more than one in_progress task, narrowing the loosely-typed args into a real TodoItem[]. Both UIs render off the existing session/event: the stdio UI prints a glyphed checklist; the ACP bridge maps the list to a `plan` sessionUpdate (todosToPlan synthesizes the priority ACP requires; status maps 1:1). Wired into the coding-agent, acp-agent, and snapshot example configs with a system-prompt nudge. Tests: unit (schema, validation, append/replace, no-agent rejection, presentCall, HMR-safety, Loader export-shape guard), full-loop integration through the agent loop, the ACP todosToPlan mapping + stream-update arm, the stdio render arm, and a session/load replay that re-emits the plan. New-group TS wiring added to tsconfig.base/json/build. RFC + a doc-inventory sweep (architecture, packages README, AGENTS layout, cookbook group list, example READMEs) ship with it. The todo-plan ACP snapshot scenario is recorded separately (needs an API key).
2026-06-29 10:30:52 +08:00
tool-todo --> agent
tool-todo --> session
tool-todo --> tools
chore: adapt web seam to master's tsconfig + regenerate catalogs Register the five web packages in the root tsconfig.json project graph (master's typecheck moved to `tsc -b tsconfig.json` and dropped the separate tsconfig.typecheck.json), and regenerate the module graph and cordis catalog so they reflect the web packages on master.
2026-06-26 19:18:46 +08:00
tool-web --> llm
tool-web --> system-prompt
tool-web --> tools
tool-web --> web
workflow: dynamic workflows — script-driven multi-agent orchestration A new capability family at packages/workflow/ in the bash seam shape, modeled on Claude Code's dynamic workflows: the model writes a JavaScript orchestration script (export const meta = {...} + plain-JS body), a runtime executes it, and the script — not the conversation — holds the loop, the branching, and the intermediate results. - dsh-workflow (ctx.workflows): abstract WorkflowService + run vocabulary (WorkflowRun whose result NEVER rejects) + observe-only workflow/* events carrying data snapshots (id + meta, never the live run), per-listener contained like subagent/*. - dsh-workflow-vm: in-process node:vm engine. Meta extraction via a string/comment-aware scanner (template interpolation rejected; literal evaluated alone in an empty timed context; statement blanked line- preservingly so stacks keep script line numbers). Hooks: agent(prompt, {label, phase, schema, model}) over ctx.subagents, parallel(), pipeline() (no cross-stage barrier), phase(), log(), args. Fatal-vs-null discipline: hook misuse (unknown/deferred options, bad arguments, unsupported schemas, tripped caps, seam start failures, cancellation) throws fatal WorkflowErrors the combinators RE-THROW — never dissolved into the per-item null reserved for child failures. Realm boundary: inbound values materialized by descriptor walks that never invoke accessors (defineProperty copies, __proto__-safe); outbound values rebuilt in-realm via the context's own JSON.parse. Determinism bans (Date.now/Math.random/argless new Date) kept so future resume support cannot break scripts. Caps and timeouts are validated Config. Every hook promise carries a no-op rejection consumer (app-boot exits on unhandled rejections). - dsh-tool-workflow: the model-facing workflow tool, synchronous like dsh-tool-subagent (start → await → try/finally dispose; abort bridged; non-completed → isError). Generic render card titled by a textual meta.name sniff. The tool description carries the authoring contract. Wired into examples/{coding-agent,acp-agent} with explicit-ask-only guidance. Coverage at every tier: unit (meta scanner, materializer incl. counting-getter and __proto__ regressions, combinator semantics, concurrency ceiling, caps, cancellation, no-unhandled-rejection abandon), integration over the real spawn stack, with-key e2e (real two-phase run + the tool through the registry pipeline), and a recorded ACP snapshot scenario (workflow-run, 1 child session). RFC: docs/rfc/implemented/feature/2026-07-05-dynamic-workflows.md (deferred work explicitly listed). AGENTS.md budget 1575 → 1590 for the new group's layout line.
2026-07-05 13:29:35 +08:00
tool-workflow --> agent
tool-workflow --> llm
tool-workflow --> tools
tool-workflow --> workflow
refactor(examples): extract the app spine into dsh-agent-core + app packages Implements docs/rfc/.../2026-06-20-extract-example-app-packages.md. Each example was thick — a hand-rolled start.ts, an infra preamble, nested base.yml/base-core.yml/acp-tail.yml includes, and a coupled front-door cluster enforced only by prose. This moves the composition into packages so each example is a thin leaf cordis.yml: pick the swappable backends, load one app package. New packages: - @deepseek-ai/dsh-agent-core (packages/core/agent-core): one bundle plugin that loads the providerless/executor-less/UI-less spine (timer + llm + sessions + system-prompt + tools + agents + invariants + tool-bash + agent-loop) via ctx.plugin(...) inside apply(), and forwards agent-loop's `agents` list as its own Config (export const Config = AgentLoop.Config, default []). - @deepseek-ai/dsh-stdio-agent (packages/ui/stdio-agent): terminal chat APP — agent-core + console logger + readline UI + a pre-created `main` agent, with a bin. The demo:echo/coding front door. - @deepseek-ai/dsh-acp-agent (packages/ui/acp-agent): ACP server APP — agent-core + JSONL persistence + the acp bridge, NO stdout logger, with a bin. The stdout-purity footgun is structurally unreachable from the leaf. Amendment to the RFC: hmr stays a LEAF cordis.yml entry, not baked into dsh-stdio-agent. hmr is a Loader-only dev plugin (throws without --expose-internals; the in-process test tier can't even import its decorator form), so a package statically importing it could never carry the per-file coverage gate. Unlike the console logger, a stray hmr is not a stdout-purity footgun, so leaving it at the leaf costs no safety. With hmr out, all three new packages carry in-process unit specs at 100%. Boot glue (Loader tail, .env load, snapshot-mode selection, stdin-dispose lifecycle) moves into each app's bin; start.ts and base.yml/base-core.yml/ acp-tail.yml are deleted. Each app package gets a keyless real-load-path test that boots through its bin + the cordis Loader (guarding the unwrapExports export-shape bug class, postmortem 0001). ACP snapshot replay stays green against the existing committed goldens (pure boot restructuring). RFC moved proposed->implemented with the amendment recorded; package/example/architecture docs and the module graph updated.
2026-06-21 12:03:44 +08:00
agent-core --> agent
agent-core --> agent-loop
agent-core --> invariants
agent-core --> llm
agent-core --> session
agent-core --> system-prompt
agent-core --> tool-bash
agent-core --> tools
feat(hooks): dsh-hooks-claude + dsh-hooks-codex bridges (hooks stack PR-F) The two bridge plugins that run a user's existing Claude Code / Codex hook config on the harness's typed interception seams, built on the shared dsh-hook-protocol library. A bridge is a faithfulness adapter, not a power tool: anything it does a native cordis plugin does more powerfully — the bridge exists only to run UNMODIFIED external hooks. - dsh-hooks-claude: CC dialect. Seven hook points (SessionStart, UserPromptSubmit, PreToolUse, PostToolUse, Stop, SubagentStart, SubagentStop), CC per-event stdin payloads, env + ${CLAUDE_PLUGIN_ROOT}/ ${CLAUDE_PROJECT_DIR} substitution, literal-or-regex matcher. - dsh-hooks-codex: Codex dialect — a deliberate subset. Five hook points, always-regex matcher, snake_case payloads (turn_id/model, no trailing newline), no env/substitution, block-only decisions. Both map the neutral merged outcome onto the seam's typed Decision and stamp an explicit {kind:'plugin'} source on injected context (so it is never mislabeled as a user prompt). Config parse-failure is contained; only command hooks run. updatedInput is logged+warned (input rewrite deferred); the Stop loop-guard is deferred (TODO). Tests: per-file 100% — config-parse unit branches + per-seam mappings end-to-end through the REAL loop + REAL bash + REAL shell scripts (scripted mock model only) + a real-Loader export-shape guard. A keyless ACP snapshot scenario (hook-prompt-block) proves a UserPromptSubmit hook blocks a prompt end-to-end (rejected turn -> ACP cancelled, hook/* events in the log); a with-key e2e (hooks.e2e.ts) proves a PreToolUse hook blocks real bash (verified on disk). The snapshot normalizer now scrubs hook/result.durationMs. RFC: docs/rfc/implemented/feature/2026-06-30-hook-bridges.md
2026-07-01 04:22:00 +08:00
hooks-claude --> agent
hooks-claude --> hook-protocol
hooks-claude --> llm
hooks-claude --> session
hooks-claude --> subagent
hooks-claude --> tools
Add the ACP subagent backend: out-of-process delegation (PR3) The first OUT-OF-PROCESS subagent backend, proving the seam generalizes past the in-process backends. @deepseek-ai/dsh-subagent-acp runs each child agent in a spawned subprocess, driven over the Agent Client Protocol as the CLIENT — the direction-inverted twin of the dsh-acp server bridge. Point the configured command at the acp-agent example and the harness talks to its own process. - Fresh process per run: start spawns, runs one ACP session (initialize → newSession → prompt), dispose kills the subprocess and awaits its exit. - Minimal client stub: advertises no fs/terminal; accumulates agent_message_chunk text as the result output; auto-answers session/request_permission by a configured policy (reject default / allow). No start-time capabilities (an out-of-process child can't enforce the parent's depth/tool-filter); ignores request.parent; injects only `subagents`. - StopReason mapping (end_turn→completed, cancelled→aborted, …); result resolves error/aborted on a child failure, never rejects (seam contract). - Security: credential-shaped ambient env vars are scrubbed; the child's own key is forwarded only via explicit config.env. A spawn-level error (ENOENT) is captured and raced against the ACP drive so a bad command settles error rather than crashing the parent. Testing designed at every tier: keyless integration drives a scripted mock ACP server subprocess (cancellation incl. the pre-newSession race and a torn-pipe-after-cancel, permission auto-answer, non-message updates, spawn failure, HMR, export shape) at 100% coverage; a with-key e2e drives the REAL acp-agent example process (PONG + real file write, verified on disk) — the harness driving itself. Snapshot coverage of an ACP child is deferred as TODO(acp-subagent-replay) (each child is its own process with its own replay). Stayed on @agentclientprotocol/sdk 0.25.1: the proposed 0.28.x bump only deprecates the stable ClientSideConnection/AgentSideConnection API this layer uses (33 sites incl. the server bridge), turning no-deprecated red across code this PR shouldn't rewrite — that fluent-API migration is its own follow-up. The backend needs nothing 0.28.x adds. This completes the subagent seam stack (PR1 interface → PR2 in-process → PR2.5 snapshot infra → PR3 ACP); the seam RFC moves to implemented/, amended.
2026-06-22 10:47:02 +08:00
subagent-acp --> agent
subagent-acp --> llm
subagent-acp --> subagent
Extract the shared in-process driver into dsh-subagent-inprocess (review feedback) The shared run driver lived inside dsh-subagent-spawn, so the spawn package carried fork-aware seeding logic and dsh-subagent-fork depended backward on dsh-subagent-spawn — the two in-process backends were not independent. Move the driver (startInProcessRun, depthOf, SubagentDepthError, InProcessRunOptions) into a new pure-library package @deepseek-ai/dsh-subagent-inprocess that registers nothing. spawn and fork now both depend only on that driver and neither knows about the other; spawn no longer re-exports it and fork no longer imports from spawn. Also wire BOTH backends in examples/coding-agent/cordis.yml (config-only): load dsh-subagent-spawn + dsh-subagent-fork + two dsh-tool-subagent instances with distinct toolNames (subagent → spawn, subagent_fork → fork), demonstrating that exposing multiple transports needs no code change.
2026-06-22 14:27:38 +08:00
subagent-inprocess --> agent
subagent-inprocess --> llm
subagent-inprocess --> session
subagent-inprocess --> subagent
subagent: implement structured output for in-process backends The seam vocabulary (SubagentStartRequest.outputSchema, SubagentResult .structured) existed but no in-process backend honored it — spawn/fork advertised outputSchema: false. This lands the missing half: - dsh-tools gains a structured-output JSON Schema subset (json-schema.ts): StructuredOutputSchema, assertSupportedOutputSchema (rejects loud outside the enforced subset, every violation listed), validateStructuredValue (path-qualified issues, total). outputSchema's seam type becomes this raw JSON-Schema subset instead of the author-facing SchemaSpec DSL — the schema travels verbatim to the model as a forced tool's parameters. - dsh-subagent-inprocess gains the shared structured runtime: one global structured_output capture tool (placeholder parameters) + a prepend:true agent/request listener doing FINAL-REQUEST enforcement (strip for plain agents, per-run schema for structured children — survives downstream request-replacing listeners) + an agent/turn-continuation veto that stops a child's turn once captured (no wasted extra model step). Lifetime is refcounted by backends (plugin lifetime) AND live runs (start→settle). - startInProcessRun drives the capture: subset asserted before the child exists, instruction appended to the child's system prompt, clean-finish nudge loop (structuredNudgeRetries, backend Config, default 1), captured value on result.structured; a clean finish without a capture settles 'error' (never a silent success with a missing field). - spawn + fork flip outputSchema: true and inject 'tools'.
2026-07-05 11:35:39 +08:00
subagent-inprocess --> tools
Add subagent capability seam: interface, mock backend, model-facing tool Introduce the `packages/subagent/` group and the abstract subagent seam — an agent delegating to a child agent — as a named-provider registry (`ctx.subagents`), unlike the single-implementation bash seam, so multiple transports (in-process, ACP, future A2A) coexist. This first PR lands the interface, a scripted test backend, and the model-facing tool, validated through the real cordis load path. - dsh-subagent: SubagentService registry + SubagentProvider/SubagentRun vocabulary + subagent/start|end events. Start-time capabilities (outputSchema, depthLimit, toolFilter) are checked pre-start and rejected loud; runtime capabilities (sendMessage, resume) are optional methods on SubagentRun. - dsh-subagent-mock (support): scripted provider for keyless, deterministic tests through the real Loader/export path. - dsh-tool-subagent: the model-facing `subagent` tool, config-bound to one provider; synchronous collect with try/finally dispose, signal->cancel bridging, and non-completed-stop-reason -> isError mapping. - Proposed RFC documenting the seam, the fork-vs-spawn-as-separate-backends decision, own-session isolation, synchronous-collect scope, and the deferral of background/poll/spill to a future unification with bash. - Wire the new group into tsconfigs, build refs, package hierarchy docs, the module graph, and the cordis catalog. RFC: docs/rfc/proposed/feature/2026-06-21-subagent-capability-seam.md
2026-06-21 22:31:56 +08:00
subagent-mock --> agent
subagent-mock --> llm
subagent-mock --> subagent
tool-subagent --> agent
tool-subagent --> llm
tool-subagent --> subagent
tool-subagent --> tools
workflow: dynamic workflows — script-driven multi-agent orchestration A new capability family at packages/workflow/ in the bash seam shape, modeled on Claude Code's dynamic workflows: the model writes a JavaScript orchestration script (export const meta = {...} + plain-JS body), a runtime executes it, and the script — not the conversation — holds the loop, the branching, and the intermediate results. - dsh-workflow (ctx.workflows): abstract WorkflowService + run vocabulary (WorkflowRun whose result NEVER rejects) + observe-only workflow/* events carrying data snapshots (id + meta, never the live run), per-listener contained like subagent/*. - dsh-workflow-vm: in-process node:vm engine. Meta extraction via a string/comment-aware scanner (template interpolation rejected; literal evaluated alone in an empty timed context; statement blanked line- preservingly so stacks keep script line numbers). Hooks: agent(prompt, {label, phase, schema, model}) over ctx.subagents, parallel(), pipeline() (no cross-stage barrier), phase(), log(), args. Fatal-vs-null discipline: hook misuse (unknown/deferred options, bad arguments, unsupported schemas, tripped caps, seam start failures, cancellation) throws fatal WorkflowErrors the combinators RE-THROW — never dissolved into the per-item null reserved for child failures. Realm boundary: inbound values materialized by descriptor walks that never invoke accessors (defineProperty copies, __proto__-safe); outbound values rebuilt in-realm via the context's own JSON.parse. Determinism bans (Date.now/Math.random/argless new Date) kept so future resume support cannot break scripts. Caps and timeouts are validated Config. Every hook promise carries a no-op rejection consumer (app-boot exits on unhandled rejections). - dsh-tool-workflow: the model-facing workflow tool, synchronous like dsh-tool-subagent (start → await → try/finally dispose; abort bridged; non-completed → isError). Generic render card titled by a textual meta.name sniff. The tool description carries the authoring contract. Wired into examples/{coding-agent,acp-agent} with explicit-ask-only guidance. Coverage at every tier: unit (meta scanner, materializer incl. counting-getter and __proto__ regressions, combinator semantics, concurrency ceiling, caps, cancellation, no-unhandled-rejection abandon), integration over the real spawn stack, with-key e2e (real two-phase run + the tool through the registry pipeline), and a recorded ACP snapshot scenario (workflow-run, 1 child session). RFC: docs/rfc/implemented/feature/2026-07-05-dynamic-workflows.md (deferred work explicitly listed). AGENTS.md budget 1575 → 1590 for the new group's layout line.
2026-07-05 13:29:35 +08:00
workflow-vm --> agent
workflow-vm --> brand
workflow-vm --> llm
workflow-vm --> subagent
workflow-vm --> tools
workflow-vm --> workflow
refactor(examples): extract the app spine into dsh-agent-core + app packages Implements docs/rfc/.../2026-06-20-extract-example-app-packages.md. Each example was thick — a hand-rolled start.ts, an infra preamble, nested base.yml/base-core.yml/acp-tail.yml includes, and a coupled front-door cluster enforced only by prose. This moves the composition into packages so each example is a thin leaf cordis.yml: pick the swappable backends, load one app package. New packages: - @deepseek-ai/dsh-agent-core (packages/core/agent-core): one bundle plugin that loads the providerless/executor-less/UI-less spine (timer + llm + sessions + system-prompt + tools + agents + invariants + tool-bash + agent-loop) via ctx.plugin(...) inside apply(), and forwards agent-loop's `agents` list as its own Config (export const Config = AgentLoop.Config, default []). - @deepseek-ai/dsh-stdio-agent (packages/ui/stdio-agent): terminal chat APP — agent-core + console logger + readline UI + a pre-created `main` agent, with a bin. The demo:echo/coding front door. - @deepseek-ai/dsh-acp-agent (packages/ui/acp-agent): ACP server APP — agent-core + JSONL persistence + the acp bridge, NO stdout logger, with a bin. The stdout-purity footgun is structurally unreachable from the leaf. Amendment to the RFC: hmr stays a LEAF cordis.yml entry, not baked into dsh-stdio-agent. hmr is a Loader-only dev plugin (throws without --expose-internals; the in-process test tier can't even import its decorator form), so a package statically importing it could never carry the per-file coverage gate. Unlike the console logger, a stray hmr is not a stdout-purity footgun, so leaving it at the leaf costs no safety. With hmr out, all three new packages carry in-process unit specs at 100%. Boot glue (Loader tail, .env load, snapshot-mode selection, stdin-dispose lifecycle) moves into each app's bin; start.ts and base.yml/base-core.yml/ acp-tail.yml are deleted. Each app package gets a keyless real-load-path test that boots through its bin + the cordis Loader (guarding the unwrapExports export-shape bug class, postmortem 0001). ACP snapshot replay stays green against the existing committed goldens (pure boot restructuring). RFC moved proposed->implemented with the amendment recorded; package/example/architecture docs and the module graph updated.
2026-06-21 12:03:44 +08:00
acp-agent --> acp
acp-agent --> agent-core
feat(ui): share the app bins' boot glue in @deepseek-ai/dsh-app-boot The four near-twin helpers the two published bins carried — loadEnv, installFailLoud, assertEntriesLoaded, boot — live once in packages/ui/app-boot, parameterized by the bin's diagnostic prefix and injectable at their side-effect seams (warn sink, process slice), so every branch sits under the per-file 100% coverage gate: the unit suite drives boot() in-process against the real Loader (relative-specifier configs) through both the settled-tree path and the fiber-less-entry rejection, and exercises the ENOENT/unloadable .env split, the Error/non-Error/stackless fail-loud arms, and the disabled-entry exclusion. resolveConfigPath (snapshot-aware) becomes the single path resolver for both bins. Each bin.ts is now a thin self-executing composition plus its app-specific lifecycle (acp: replay env-skip + stdin-EOF dispose; stdio: nothing extra), exports nothing, and stays coverage-excluded; the built-bin smokes still prove both artifacts under plain node in the node_modules-shaped temp dir (now symlinking ui/app-boot), including the missing-config non-zero exit. Implements docs/rfc/implemented/simplification/2026-07-04-share-app-bin-boot-glue.md (moved from proposed/ and amended); the extract-example-app-packages RFC's bin-ownership facts are amended in the same change.
2026-07-04 16:43:06 +08:00
acp-agent --> app-boot
refactor(examples): extract the app spine into dsh-agent-core + app packages Implements docs/rfc/.../2026-06-20-extract-example-app-packages.md. Each example was thick — a hand-rolled start.ts, an infra preamble, nested base.yml/base-core.yml/acp-tail.yml includes, and a coupled front-door cluster enforced only by prose. This moves the composition into packages so each example is a thin leaf cordis.yml: pick the swappable backends, load one app package. New packages: - @deepseek-ai/dsh-agent-core (packages/core/agent-core): one bundle plugin that loads the providerless/executor-less/UI-less spine (timer + llm + sessions + system-prompt + tools + agents + invariants + tool-bash + agent-loop) via ctx.plugin(...) inside apply(), and forwards agent-loop's `agents` list as its own Config (export const Config = AgentLoop.Config, default []). - @deepseek-ai/dsh-stdio-agent (packages/ui/stdio-agent): terminal chat APP — agent-core + console logger + readline UI + a pre-created `main` agent, with a bin. The demo:echo/coding front door. - @deepseek-ai/dsh-acp-agent (packages/ui/acp-agent): ACP server APP — agent-core + JSONL persistence + the acp bridge, NO stdout logger, with a bin. The stdout-purity footgun is structurally unreachable from the leaf. Amendment to the RFC: hmr stays a LEAF cordis.yml entry, not baked into dsh-stdio-agent. hmr is a Loader-only dev plugin (throws without --expose-internals; the in-process test tier can't even import its decorator form), so a package statically importing it could never carry the per-file coverage gate. Unlike the console logger, a stray hmr is not a stdout-purity footgun, so leaving it at the leaf costs no safety. With hmr out, all three new packages carry in-process unit specs at 100%. Boot glue (Loader tail, .env load, snapshot-mode selection, stdin-dispose lifecycle) moves into each app's bin; start.ts and base.yml/base-core.yml/ acp-tail.yml are deleted. Each app package gets a keyless real-load-path test that boots through its bin + the cordis Loader (guarding the unwrapExports export-shape bug class, postmortem 0001). ACP snapshot replay stays green against the existing committed goldens (pure boot restructuring). RFC moved proposed->implemented with the amendment recorded; package/example/architecture docs and the module graph updated.
2026-06-21 12:03:44 +08:00
acp-agent --> session-persistence-jsonl
stdio-agent --> agent
stdio-agent --> agent-core
feat(ui): share the app bins' boot glue in @deepseek-ai/dsh-app-boot The four near-twin helpers the two published bins carried — loadEnv, installFailLoud, assertEntriesLoaded, boot — live once in packages/ui/app-boot, parameterized by the bin's diagnostic prefix and injectable at their side-effect seams (warn sink, process slice), so every branch sits under the per-file 100% coverage gate: the unit suite drives boot() in-process against the real Loader (relative-specifier configs) through both the settled-tree path and the fiber-less-entry rejection, and exercises the ENOENT/unloadable .env split, the Error/non-Error/stackless fail-loud arms, and the disabled-entry exclusion. resolveConfigPath (snapshot-aware) becomes the single path resolver for both bins. Each bin.ts is now a thin self-executing composition plus its app-specific lifecycle (acp: replay env-skip + stdin-EOF dispose; stdio: nothing extra), exports nothing, and stays coverage-excluded; the built-bin smokes still prove both artifacts under plain node in the node_modules-shaped temp dir (now symlinking ui/app-boot), including the missing-config non-zero exit. Implements docs/rfc/implemented/simplification/2026-07-04-share-app-bin-boot-glue.md (moved from proposed/ and amended); the extract-example-app-packages RFC's bin-ownership facts are amended in the same change.
2026-07-04 16:43:06 +08:00
stdio-agent --> app-boot
docs: regenerate the module graph and dependency row for the dsh-llm edge
2026-07-04 17:17:36 +08:00
stdio-agent --> llm
refactor(examples): extract the app spine into dsh-agent-core + app packages Implements docs/rfc/.../2026-06-20-extract-example-app-packages.md. Each example was thick — a hand-rolled start.ts, an infra preamble, nested base.yml/base-core.yml/acp-tail.yml includes, and a coupled front-door cluster enforced only by prose. This moves the composition into packages so each example is a thin leaf cordis.yml: pick the swappable backends, load one app package. New packages: - @deepseek-ai/dsh-agent-core (packages/core/agent-core): one bundle plugin that loads the providerless/executor-less/UI-less spine (timer + llm + sessions + system-prompt + tools + agents + invariants + tool-bash + agent-loop) via ctx.plugin(...) inside apply(), and forwards agent-loop's `agents` list as its own Config (export const Config = AgentLoop.Config, default []). - @deepseek-ai/dsh-stdio-agent (packages/ui/stdio-agent): terminal chat APP — agent-core + console logger + readline UI + a pre-created `main` agent, with a bin. The demo:echo/coding front door. - @deepseek-ai/dsh-acp-agent (packages/ui/acp-agent): ACP server APP — agent-core + JSONL persistence + the acp bridge, NO stdout logger, with a bin. The stdout-purity footgun is structurally unreachable from the leaf. Amendment to the RFC: hmr stays a LEAF cordis.yml entry, not baked into dsh-stdio-agent. hmr is a Loader-only dev plugin (throws without --expose-internals; the in-process test tier can't even import its decorator form), so a package statically importing it could never carry the per-file coverage gate. Unlike the console logger, a stray hmr is not a stdout-purity footgun, so leaving it at the leaf costs no safety. With hmr out, all three new packages carry in-process unit specs at 100%. Boot glue (Loader tail, .env load, snapshot-mode selection, stdin-dispose lifecycle) moves into each app's bin; start.ts and base.yml/base-core.yml/ acp-tail.yml are deleted. Each app package gets a keyless real-load-path test that boots through its bin + the cordis Loader (guarding the unwrapExports export-shape bug class, postmortem 0001). ACP snapshot replay stays green against the existing committed goldens (pure boot restructuring). RFC moved proposed->implemented with the amendment recorded; package/example/architecture docs and the module graph updated.
2026-06-21 12:03:44 +08:00
stdio-agent --> session
stdio-agent --> session-persistence-jsonl
Add in-process subagent backends: spawn (fresh) and fork (seeded) The second PR of the subagent seam: the two in-process backends that run a child agent on the same cordis context, reusing the agent factory's quiescent AgentHandle teardown. Both register on ctx.subagents (PR1's named-provider registry) and share one run driver. - dsh-subagent-spawn: a FRESH child via ctx.agents.create — own session, the parent's model by default (overridable), zero inherited conversation. Also exports the shared in-process run driver (startInProcessRun): mint ids, stamp cwd/parentSession-lineage/depth, drive the one-shot (send → whenIdle), read the last assistant/message + turn/end reason, dispose to quiescence. - dsh-subagent-fork: a child SEEDED with the parent's balanced completed-turn prefix (the log up to and including its last turn/end), so the child inherits context. The in-flight unbalanced turn is excluded — a raw seed would fail the invariants replay. Proven: a regression test goes red if the boundary seeds the open turn. - Seam extension: CreateAgentOptions.seed, threaded through AgentLoop.createAgent → ctx.sessions.prepare({ seed }) (the primitive resume already used). This is the fork-lineage path the TODO(sub-agents) markers anticipated. - Depth: a merge-extensible AgentOptions.subagentDepth (0 top-level, parent+1 for a child); the depthLimit capability refuses a spawn past request.maxDepth. Tests: real-loop unit tests for both backends (mock MODEL only, real loop + invariants), a multi-subagent test (one parent drives a fork AND a spawn child then keeps working), and a with-key e2e (a real parent delegates via the `subagent` tool to a real child that writes a file on disk — world-verified). 100% per-file coverage. The coding-agent demo wires the spawn backend + tool. Snapshot coverage of nested agents is deferred to a stacked follow-up (TODO(subagent-snapshots)): dsh-llm-replay is a single global positional cursor that cannot route calls to a parent vs. a child on one context. Recorded in the RFC's deferrals and a new AGENTS.md rule: designing a subsystem must design its test infrastructure END TO END up front, verifying the snapshot/e2e harness can express the new shape — a gap this plan hit.
2026-06-22 05:58:40 +08:00
subagent-fork --> agent
subagent-fork --> session
subagent-fork --> subagent
Extract the shared in-process driver into dsh-subagent-inprocess (review feedback) The shared run driver lived inside dsh-subagent-spawn, so the spawn package carried fork-aware seeding logic and dsh-subagent-fork depended backward on dsh-subagent-spawn — the two in-process backends were not independent. Move the driver (startInProcessRun, depthOf, SubagentDepthError, InProcessRunOptions) into a new pure-library package @deepseek-ai/dsh-subagent-inprocess that registers nothing. spawn and fork now both depend only on that driver and neither knows about the other; spawn no longer re-exports it and fork no longer imports from spawn. Also wire BOTH backends in examples/coding-agent/cordis.yml (config-only): load dsh-subagent-spawn + dsh-subagent-fork + two dsh-tool-subagent instances with distinct toolNames (subagent → spawn, subagent_fork → fork), demonstrating that exposing multiple transports needs no code change.
2026-06-22 14:27:38 +08:00
subagent-fork --> subagent-inprocess
subagent-spawn --> subagent
subagent-spawn --> subagent-inprocess
feat: generate module dependency graph with freshness gate Add scripts/gen-module-graph.ts, which derives the inter-package dependency graph from each package's @deepseek-ai/dsh-* peerDependencies and renders docs/module-graph.md (a GitHub-native Mermaid graph plus a dependency table). Output is deterministic so a regenerate-and-diff check is stable. Wire a freshness gate the same way doc-sync is wired (ADR 0007: hooks and CI run the same package.json scripts): verify-module-graph runs in pre-push (lefthook) and as a CI step. It fails if the committed file drifts from what the generator would produce.
2026-06-16 21:00:24 +08:00
```
| Package | Depends on |
| --- | --- |
feat(ui): share the app bins' boot glue in @deepseek-ai/dsh-app-boot The four near-twin helpers the two published bins carried — loadEnv, installFailLoud, assertEntriesLoaded, boot — live once in packages/ui/app-boot, parameterized by the bin's diagnostic prefix and injectable at their side-effect seams (warn sink, process slice), so every branch sits under the per-file 100% coverage gate: the unit suite drives boot() in-process against the real Loader (relative-specifier configs) through both the settled-tree path and the fiber-less-entry rejection, and exercises the ENOENT/unloadable .env split, the Error/non-Error/stackless fail-loud arms, and the disabled-entry exclusion. resolveConfigPath (snapshot-aware) becomes the single path resolver for both bins. Each bin.ts is now a thin self-executing composition plus its app-specific lifecycle (acp: replay env-skip + stdin-EOF dispose; stdio: nothing extra), exports nothing, and stays coverage-excluded; the built-bin smokes still prove both artifacts under plain node in the node_modules-shaped temp dir (now symlinking ui/app-boot), including the missing-config non-zero exit. Implements docs/rfc/implemented/simplification/2026-07-04-share-app-bin-boot-glue.md (moved from proposed/ and amended); the extract-example-app-packages RFC's bin-ownership facts are amended in the same change.
2026-07-04 16:43:06 +08:00
| `app-boot` | — |
feat(types): brand bash ids + stop brand erosion; extract Branded to dsh-brand Type-only change (brands are zero-cost casts; no runtime/wire impact). Closes the two gaps in the "brand ids that cross package boundaries" policy and fixes the dependency direction so a capability package never pulls in an unrelated one. - Extract the `Branded<B>` primitive into a new standalone type-only package `@deepseek-ai/dsh-brand` (packages/util/brand) with no harness-package deps. dsh-llm keeps its owned CallId but imports Branded from dsh-brand; dsh-session, dsh-agent, and dsh-bash all import Branded from there. dsh-bash depends on dsh-brand ALONE — never on dsh-llm or dsh-session (the architectural fix: a generic execution backend must not couple to the LLM or session vocabulary). - Mint BashTaskId + OwnerToken in dsh-bash and thread them through BashTask.id, the get/ownerOf/list/readOutput/kill seam, the bash-local generation site, and the dsh-tool-bash validate/access surface. OwnerToken is a DISTINCT brand from SessionId so the seam stays decoupled; dsh-tool-bash is the single boundary that casts SessionId -> OwnerToken. - Brand at the SOURCE, not via mid-pipeline casts: agent-loop's Config types agents[].id as AgentId and resumeSessionId as SessionId, so the brand enters at the config boundary and the inner create()/resume casts disappear (only the genuinely-new per-run session-id string is cast). - Stop brand erosion: propagate CallId/SessionId/AgentId to the registry/store Map keys and public params/exports (SessionStore, AgentRegistry + factory options, the ACP session-id surface + ToolPresenter CallId map, the persistence coordinator, invariants pendingCalls, the pi-ai tool-call maps). - Docs: document BashTaskId/OwnerToken in bash.md (type-equiv re-pasted), point the Branded type-equiv at dsh-brand, fix stale param types in the session/ agent/bash READMEs, regenerate the cordis catalog + module graph. Implements docs/rfc/proposed/architecture/2026-06-20-branded-ids.md
2026-06-21 07:17:25 +08:00
| `brand` | — |
| `bash` | `brand` |
| `llm` | `brand` |
refactor: order module-graph table topologically (low→high level) Replace the alphabetical package ordering with a Kahn-style topological sort (alphabetical tiebreak within each layer to stay deterministic), so the table and graph list leaf interfaces first and dependents last.
2026-06-16 21:11:09 +08:00
| `bash-local` | `bash` |
refactor(fs): split filesystem seam into provider ctx.fs + policy ctx.fileContext Implements the split-the-filesystem-seam RFC. ctx.fs shrinks to a text-storage provider seam (resolve/stat/readText/streamText/writeText/editText with branded FsTargetKey/FsVersion and an explicit FsWriteExpectation); the new dsh-file-context package owns the model-facing policy (read windowing, observed-state, write/edit freshness) as the concrete ctx.fileContext service. Authorization is now freshness-based rather than full/partial view: a windowed read records the file version and authorizes a later edit when the file is unchanged, removing the dead-end where reading lines 100-150 of a large file could not edit line 120. editText stays a provider primitive so version guard + literal match + atomic rewrite remain one critical section, and the stale check runs before matching so a stale edit reports FS_STALE_VERSION. tool-fs injects fileContext, never reaching around to ctx.fs (the no-bypass contract).
2026-06-26 17:23:18 +08:00
| `fs` | `brand`, `llm` |
feat: generate module dependency graph with freshness gate Add scripts/gen-module-graph.ts, which derives the inter-package dependency graph from each package's @deepseek-ai/dsh-* peerDependencies and renders docs/module-graph.md (a GitHub-native Mermaid graph plus a dependency table). Output is deterministic so a regenerate-and-diff check is stable. Wire a freshness gate the same way doc-sync is wired (ADR 0007: hooks and CI run the same package.json scripts): verify-module-graph runs in pre-push (lefthook) and as a CI step. It fails if the committed file drifts from what the generator would produce.
2026-06-16 21:00:24 +08:00
| `llm-deepseek` | `llm` |
| `llm-pi-ai` | `llm` |
feat(types): brand bash ids + stop brand erosion; extract Branded to dsh-brand Type-only change (brands are zero-cost casts; no runtime/wire impact). Closes the two gaps in the "brand ids that cross package boundaries" policy and fixes the dependency direction so a capability package never pulls in an unrelated one. - Extract the `Branded<B>` primitive into a new standalone type-only package `@deepseek-ai/dsh-brand` (packages/util/brand) with no harness-package deps. dsh-llm keeps its owned CallId but imports Branded from dsh-brand; dsh-session, dsh-agent, and dsh-bash all import Branded from there. dsh-bash depends on dsh-brand ALONE — never on dsh-llm or dsh-session (the architectural fix: a generic execution backend must not couple to the LLM or session vocabulary). - Mint BashTaskId + OwnerToken in dsh-bash and thread them through BashTask.id, the get/ownerOf/list/readOutput/kill seam, the bash-local generation site, and the dsh-tool-bash validate/access surface. OwnerToken is a DISTINCT brand from SessionId so the seam stays decoupled; dsh-tool-bash is the single boundary that casts SessionId -> OwnerToken. - Brand at the SOURCE, not via mid-pipeline casts: agent-loop's Config types agents[].id as AgentId and resumeSessionId as SessionId, so the brand enters at the config boundary and the inner create()/resume casts disappear (only the genuinely-new per-run session-id string is cast). - Stop brand erosion: propagate CallId/SessionId/AgentId to the registry/store Map keys and public params/exports (SessionStore, AgentRegistry + factory options, the ACP session-id surface + ToolPresenter CallId map, the persistence coordinator, invariants pendingCalls, the pi-ai tool-call maps). - Docs: document BashTaskId/OwnerToken in bash.md (type-equiv re-pasted), point the Branded type-equiv at dsh-brand, fix stale param types in the session/ agent/bash READMEs, regenerate the cordis catalog + module graph. Implements docs/rfc/proposed/architecture/2026-06-20-branded-ids.md
2026-06-21 07:17:25 +08:00
| `session` | `brand`, `llm` |
feat: generate module dependency graph with freshness gate Add scripts/gen-module-graph.ts, which derives the inter-package dependency graph from each package's @deepseek-ai/dsh-* peerDependencies and renders docs/module-graph.md (a GitHub-native Mermaid graph plus a dependency table). Output is deterministic so a regenerate-and-diff check is stable. Wire a freshness gate the same way doc-sync is wired (ADR 0007: hooks and CI run the same package.json scripts): verify-module-graph runs in pre-push (lefthook) and as a CI step. It fails if the committed file drifts from what the generator would produce.
2026-06-16 21:00:24 +08:00
| `system-prompt` | `llm` |
chore: adapt web seam to master's tsconfig + regenerate catalogs Register the five web packages in the root tsconfig.json project graph (master's typecheck moved to `tsc -b tsconfig.json` and dropped the separate tsconfig.typecheck.json), and regenerate the module graph and cordis catalog so they reflect the web packages on master.
2026-06-26 19:18:46 +08:00
| `web` | `llm` |
feat(types): brand bash ids + stop brand erosion; extract Branded to dsh-brand Type-only change (brands are zero-cost casts; no runtime/wire impact). Closes the two gaps in the "brand ids that cross package boundaries" policy and fixes the dependency direction so a capability package never pulls in an unrelated one. - Extract the `Branded<B>` primitive into a new standalone type-only package `@deepseek-ai/dsh-brand` (packages/util/brand) with no harness-package deps. dsh-llm keeps its owned CallId but imports Branded from dsh-brand; dsh-session, dsh-agent, and dsh-bash all import Branded from there. dsh-bash depends on dsh-brand ALONE — never on dsh-llm or dsh-session (the architectural fix: a generic execution backend must not couple to the LLM or session vocabulary). - Mint BashTaskId + OwnerToken in dsh-bash and thread them through BashTask.id, the get/ownerOf/list/readOutput/kill seam, the bash-local generation site, and the dsh-tool-bash validate/access surface. OwnerToken is a DISTINCT brand from SessionId so the seam stays decoupled; dsh-tool-bash is the single boundary that casts SessionId -> OwnerToken. - Brand at the SOURCE, not via mid-pipeline casts: agent-loop's Config types agents[].id as AgentId and resumeSessionId as SessionId, so the brand enters at the config boundary and the inner create()/resume casts disappear (only the genuinely-new per-run session-id string is cast). - Stop brand erosion: propagate CallId/SessionId/AgentId to the registry/store Map keys and public params/exports (SessionStore, AgentRegistry + factory options, the ACP session-id surface + ToolPresenter CallId map, the persistence coordinator, invariants pendingCalls, the pi-ai tool-call maps). - Docs: document BashTaskId/OwnerToken in bash.md (type-equiv re-pasted), point the Branded type-equiv at dsh-brand, fix stale param types in the session/ agent/bash READMEs, regenerate the cordis catalog + module graph. Implements docs/rfc/proposed/architecture/2026-06-20-branded-ids.md
2026-06-21 07:17:25 +08:00
| `agent` | `brand`, `llm`, `session` |
feat(compact): compaction capability seam — abstract CompactService interface Adds the @deepseek-ai/dsh-compact interface package: the abstract CompactService (ctx.compact) with compactIfNeeded / compactRegion, the compact/* session-event types via SessionEventMap declaration merging, and the capability-seam RFC. Wires the package into the three root tsconfigs and the cordis catalog. A backend implementation lands separately.
2026-06-22 14:52:30 +08:00
| `compact` | `llm`, `session` |
Add filesystem capability seam and tools
2026-06-22 10:48:41 +08:00
| `fs-local` | `fs` |
fix(fs): address review — rename to dsh-fs-policy, fs/*-intent events, RFC currency, ENOTDIR Rename per review naming decisions: - package dsh-file-context → dsh-fs-policy (dir, package name, plugin name, tsconfig refs, importers, type-equiv manifest, generated catalog + module-graph) - events fs/write-expectation → fs/write-intent, fs/edit-expectation → fs/edit-intent (fs/observed unchanged); type FsWriteExpectation → FsWriteIntent, "expectation" wording → "intent" throughout - exported FileContextExec → FsPolicyExec Make the implemented RFCs describe what shipped, not the superseded designs: the 2026-06-17 capability-seam + tool-schemas RFCs no longer place policy on ctx.fs or use full/partial-view authorization, and the fsspec RFC's ctx.fileContext service prose is rewritten to the fs/* event-gate reality (freshness-based auth). Sharpen docs/rfc/implemented/AGENTS.md: a rename is a fact to fix IN PLACE — the "new RFC" escape hatch is for macro decision reversals only, not renames. Code fixes from review: - fsio.ts resolveLocalTarget/probe translate ENOTDIR (a parent path segment is a file) into the structured FsError taxonomy instead of leaking a raw Node error; resolve reports FS_NOT_FOUND, probe reports absent. Regression tests proven to fail on the unfixed code. - tool-fs HMR test now asserts prompt sections (not just tool schemas) are withdrawn on disposal. - fs/observed is a plain (unguarded) ctx.emit: correct the fs-policy comment, filesystem.md, and tool-fs module doc that wrongly claimed the tool "contains" a throwing listener; a throw surfaces as the tool's isError result. - drop the false "loaded by the default product config" claim (no config wires the fs tools yet), the duplicate ctx.bash service-map row, the stale FileReadRequest catalog link-map entry, and the fs/fs README EOF blank line; correct the dsh-fs package.json description.
2026-07-02 03:12:38 +08:00
| `fs-policy` | `fs` |
feat(hooks): dsh-hook-protocol — shared Claude Code / Codex hook wire-protocol core The two hook bridges (dsh-hooks-claude, dsh-hooks-codex) would otherwise duplicate the bulk of the protocol — Codex deliberately reimplements a SUBSET of the Claude Code protocol (same hooks.json shape, exit-code/stdout contract, command-hook model). This library holds the genuinely-identical primitives; each bridge owns only what differs (per-event stdin payload, env/substitution, decision mapping). New packages/hooks/ group; hook-protocol is a LIBRARY (no plugin, registers/injects nothing): - matcher: matchesMatcher(pattern, query, mode) — the one dialect axis collapsed to a mode param (claude = literal-or-regex with pipe alternation; codex = always unanchored regex). Match-all on absent/''/'*'; invalid regex matches nothing. - codec: parseHookOutput(exit, stdout, stderr) → dialect-neutral HookOutput. Exit 0 → lenient JSON; exit 2 → blocking error (stderr = reason, surfaced as decision:'block'); other → non-blocking. Parses the CC superset (continue/stopReason/decision/hookSpecificOutput.{permissionDecision, additionalContext,updatedInput}/systemMessage); permissionDecision overrides the legacy top-level decision. - runner: runHook(bash, hook, opts, now) — runs a command hook via ctx.bash (stdin payload + trusted-plugin env), honors timeoutSec, never throws (executor reject → non-blocking-error HookOutput). Injected clock for testable durations. - merge: mergeHookOutputs — most-restrictive fold (deny>ask>allow, sticky stop, block reasons joined, context/system-messages accumulated). - hook/* session events (declaration-merged into SessionEventMap, log-only like compact/*) + appendHookInvoked/appendHookResult helpers. updatedInput is parsed but NOT honored (deferred pre-tool-input-rewrite RFC); a bridge logs+warns. 47 unit tests at per-file 100% (matcher per-mode, codec per exit-code/field, runner plumbing w/ stub executor, merge precedence, hook/* helpers). RFC: implemented/feature/2026-06-30-hook-protocol-lib.md.
2026-07-01 00:38:06 +08:00
| `hook-protocol` | `bash`, `session` |
refactor(examples): extract reusable logic into tested packages Logic that lived under examples/ was outside the per-file 100% coverage gate (examples/ are not workspaces) and, in the stdio-UI case, duplicated across two examples. Move it into packages/ so it is gated and de-duped. - packages/ui-stdio (new): unify the two diverged stdio-chat.ts copies into one @deepseek-ai/dsh-ui-stdio plugin (welcome/agent Config). A test-only I/O seam (createStdioChat(ctx, config, runtime)) keeps process streams out of the serializable config and makes every render/EOF/disposal branch unit-testable. Per-file 100%. echo/coding cordis.yml now load the package; both src/stdio-chat.ts deleted. - packages/llm-replay (new): move examples/acp-agent/src/llm-replay.ts (+ its spec) here so its derive/parse/replay branches fall under the coverage gate. cordis.snapshot.yml + README rewired to the package name; added apply/env /assertNever/abort tests to reach per-file 100%. - examples/{echo,coding}-agent: keyless Loader-path e2e smokes that boot the real cordis.yml (no key) — the guard a hand-mounted unit test cannot be for the unwrapExports/export-shape class (postmortem 0001). examples/AGENTS.md codifies the keyless+with-key smoke convention (keyless-by-nature exception for echo-agent). - AGENTS.md: a scoped, removal-triggered pre-release stance (foundation over blast radius). packages/README.md: new rows + a FIXME to later regroup ALL packages into a hierarchy. Wiring: tsconfig paths/refs, publint, knip, module-graph. Verified: typecheck, lint, test:coverage (887 tests, 100%), build, hygiene, doc-sync, test:snapshot (10), test:e2e (6 keyless pass, with-key self-skip).
2026-06-19 12:42:28 +08:00
| `llm-replay` | `llm`, `session` |
Merge remote-tracking branch 'origin/master' into split/session-persistence
2026-06-16 23:16:59 +08:00
| `session-persistence` | `session` |
chore: adapt web seam to master's tsconfig + regenerate catalogs Register the five web packages in the root tsconfig.json project graph (master's typecheck moved to `tsc -b tsconfig.json` and dropped the separate tsconfig.typecheck.json), and regenerate the module graph and cordis catalog so they reflect the web packages on master.
2026-06-26 19:18:46 +08:00
| `web-fetch-local` | `web` |
feat(web): add DeepSeek-backed web search provider Add @deepseek-ai/dsh-web-search-deepseek: a WebSearchProvider that calls DeepSeek's Anthropic-compatible Messages API with the native web_search_20250305 server tool and parses the structured web_search_tool_result blocks into the ctx.web seam's WebSearchResult. - Namespace plugin (inject: ['web']), no default export — registers into ctx.web like dsh-llm-deepseek registers into ctx.llm. - Strict mode: a response with no web_search_tool_result block throws WEB_PROVIDER_ERROR rather than scraping URLs from model prose. - Reuses $DEEPSEEK_API_KEY; baseURL defaults to the Anthropic-compatible base (api.deepseek.com/anthropic/v1) and does NOT reuse $DEEPSEEK_BASE_URL, which belongs to the chat-completions LLM adapter. - snippet joined from text-block citations; sources deduped by url. - Two-stage build layout (outDir lib/types) matching the other web packages; registered in tsconfig.json, tsconfig.build.json, knip.json, and docs/module-graph.md.
2026-06-29 15:21:12 +08:00
| `web-search-deepseek` | `web` |
chore: adapt web seam to master's tsconfig + regenerate catalogs Register the five web packages in the root tsconfig.json project graph (master's typecheck moved to `tsc -b tsconfig.json` and dropped the separate tsconfig.typecheck.json), and regenerate the module graph and cordis catalog so they reflect the web packages on master.
2026-06-26 19:18:46 +08:00
| `web-search-exa` | `web` |
| `web-search-perplexity` | `web` |
feat(compact-basic): baseline compaction backend (squashed from compact-basic) Collapses the per-round review churn of the prior compact-basic branch into a single clean baseline on top of compact-interface, so the upcoming retention refactor lands as fresh, well-scoped commits rather than stacking on a history of fixes that are being superseded.
2026-06-25 17:30:42 +08:00
| `compact-basic` | `agent`, `compact`, `llm`, `session` |
refactor: order module-graph table topologically (low→high level) Replace the alphabetical package ordering with a Kahn-style topological sort (alphabetical tiebreak within each layer to stay deterministic), so the table and graph list leaf interfaces first and dependents last.
2026-06-16 21:11:09 +08:00
| `invariants` | `agent`, `llm`, `session` |
Merge remote-tracking branch 'origin/master' into split/session-persistence
2026-06-16 23:16:59 +08:00
| `session-persistence-jsonl` | `session`, `session-persistence` |
Merge remote-tracking branch 'origin/split/agent-factory' into split/session-persistence-sqlite
2026-06-16 23:19:45 +08:00
| `session-persistence-sqlite` | `session`, `session-persistence` |
feat: generate module dependency graph with freshness gate Add scripts/gen-module-graph.ts, which derives the inter-package dependency graph from each package's @deepseek-ai/dsh-* peerDependencies and renders docs/module-graph.md (a GitHub-native Mermaid graph plus a dependency table). Output is deterministic so a regenerate-and-diff check is stable. Wire a freshness gate the same way doc-sync is wired (ADR 0007: hooks and CI run the same package.json scripts): verify-module-graph runs in pre-push (lefthook) and as a CI step. It fails if the committed file drifts from what the generator would produce.
2026-06-16 21:00:24 +08:00
| `tools` | `agent`, `llm`, `system-prompt` |
workflow: dynamic workflows — script-driven multi-agent orchestration A new capability family at packages/workflow/ in the bash seam shape, modeled on Claude Code's dynamic workflows: the model writes a JavaScript orchestration script (export const meta = {...} + plain-JS body), a runtime executes it, and the script — not the conversation — holds the loop, the branching, and the intermediate results. - dsh-workflow (ctx.workflows): abstract WorkflowService + run vocabulary (WorkflowRun whose result NEVER rejects) + observe-only workflow/* events carrying data snapshots (id + meta, never the live run), per-listener contained like subagent/*. - dsh-workflow-vm: in-process node:vm engine. Meta extraction via a string/comment-aware scanner (template interpolation rejected; literal evaluated alone in an empty timed context; statement blanked line- preservingly so stacks keep script line numbers). Hooks: agent(prompt, {label, phase, schema, model}) over ctx.subagents, parallel(), pipeline() (no cross-stage barrier), phase(), log(), args. Fatal-vs-null discipline: hook misuse (unknown/deferred options, bad arguments, unsupported schemas, tripped caps, seam start failures, cancellation) throws fatal WorkflowErrors the combinators RE-THROW — never dissolved into the per-item null reserved for child failures. Realm boundary: inbound values materialized by descriptor walks that never invoke accessors (defineProperty copies, __proto__-safe); outbound values rebuilt in-realm via the context's own JSON.parse. Determinism bans (Date.now/Math.random/argless new Date) kept so future resume support cannot break scripts. Caps and timeouts are validated Config. Every hook promise carries a no-op rejection consumer (app-boot exits on unhandled rejections). - dsh-tool-workflow: the model-facing workflow tool, synchronous like dsh-tool-subagent (start → await → try/finally dispose; abort bridged; non-completed → isError). Generic render card titled by a textual meta.name sniff. The tool description carries the authoring contract. Wired into examples/{coding-agent,acp-agent} with explicit-ask-only guidance. Coverage at every tier: unit (meta scanner, materializer incl. counting-getter and __proto__ regressions, combinator semantics, concurrency ceiling, caps, cancellation, no-unhandled-rejection abandon), integration over the real spawn stack, with-key e2e (real two-phase run + the tool through the registry pipeline), and a recorded ACP snapshot scenario (workflow-run, 1 child session). RFC: docs/rfc/implemented/feature/2026-07-05-dynamic-workflows.md (deferred work explicitly listed). AGENTS.md budget 1575 → 1590 for the new group's layout line.
2026-07-05 13:29:35 +08:00
| `workflow` | `agent`, `brand`, `llm` |
feat(acp): tool-owned tool-call UI presentation (title/command/output) In Zed the tool-call card showed only "bash" — the bare tool name — instead of what the command does. Fix it by letting each TOOL own how its calls render, rather than the bridge special-casing names. dsh-tools: add an optional two-state presentation seam to ToolDefinition / defineTool — `presentCall(args)` (pending: title, kind, rawInput) and `presentResult(args, result)` (completed: title?, content?). Provider-neutral `ToolCallKind`/`ToolCallPresentation`/`ToolResultPresentation` vocabulary so tools never depend on ACP. defineTool soft-validates args (display runs on log replay, so a malformed/old shape returns undefined instead of throwing). dsh-tool-bash: bash declares presentCall (model `description` → title, exact `command` → rawInput, kind execute) and presentResult (wrap output in a fenced ```console block — a UI-only affordance kept out of the model-facing result); bash_output/bash_kill present task-scoped titles. dsh-acp: inject `tools`; a per-session `ToolPresenter` looks the tool up by name and maps its neutral presentation to the ACP tool_call/tool_call_update wire shape, with a generic fallback (title = name) for tools that declare nothing. Because the `tool/result` event carries only {callId, content, isError}, the presenter keeps a small bridge-local map of ONLY in-flight calls' (name, args), keyed by callId and removed as each result is presented — no event-schema or core change. Replay uses a throwaway presenter so loaded sessions render identically to live ones. Tests: dsh-tools defineTool presenters (typed args, soft-validate), tool-bash bash/bash_output/bash_kill presenters, acp ToolPresenter (tool-owned mapping, unknown-callId fallback, in-flight-only map), and an end-to-end turn through the bridge. The key-gated e2e now asserts a real bash call's title is the model description (not "bash") and rawInput is the command — verified against the real DeepSeek model. The test harness derives its inject from the bridge's exported `inject` so it can't drift again.
2026-06-18 09:01:36 +08:00
| `acp` | `agent`, `llm`, `session`, `session-persistence`, `tools` |
Merge remote-tracking branch 'origin/split/session-persistence' into split/agent-factory
2026-06-16 23:18:31 +08:00
| `agent-loop` | `agent`, `llm`, `session`, `session-persistence`, `system-prompt`, `tools` |
feat(hooks): dsh-hooks-claude + dsh-hooks-codex bridges (hooks stack PR-F) The two bridge plugins that run a user's existing Claude Code / Codex hook config on the harness's typed interception seams, built on the shared dsh-hook-protocol library. A bridge is a faithfulness adapter, not a power tool: anything it does a native cordis plugin does more powerfully — the bridge exists only to run UNMODIFIED external hooks. - dsh-hooks-claude: CC dialect. Seven hook points (SessionStart, UserPromptSubmit, PreToolUse, PostToolUse, Stop, SubagentStart, SubagentStop), CC per-event stdin payloads, env + ${CLAUDE_PLUGIN_ROOT}/ ${CLAUDE_PROJECT_DIR} substitution, literal-or-regex matcher. - dsh-hooks-codex: Codex dialect — a deliberate subset. Five hook points, always-regex matcher, snake_case payloads (turn_id/model, no trailing newline), no env/substitution, block-only decisions. Both map the neutral merged outcome onto the seam's typed Decision and stamp an explicit {kind:'plugin'} source on injected context (so it is never mislabeled as a user prompt). Config parse-failure is contained; only command hooks run. updatedInput is logged+warned (input rewrite deferred); the Stop loop-guard is deferred (TODO). Tests: per-file 100% — config-parse unit branches + per-seam mappings end-to-end through the REAL loop + REAL bash + REAL shell scripts (scripted mock model only) + a real-Loader export-shape guard. A keyless ACP snapshot scenario (hook-prompt-block) proves a UserPromptSubmit hook blocks a prompt end-to-end (rejected turn -> ACP cancelled, hook/* events in the log); a with-key e2e (hooks.e2e.ts) proves a PreToolUse hook blocks real bash (verified on disk). The snapshot normalizer now scrubs hook/result.durationMs. RFC: docs/rfc/implemented/feature/2026-06-30-hook-bridges.md
2026-07-01 04:22:00 +08:00
| `hooks-codex` | `agent`, `hook-protocol`, `llm`, `session`, `tools` |
Add subagent capability seam: interface, mock backend, model-facing tool Introduce the `packages/subagent/` group and the abstract subagent seam — an agent delegating to a child agent — as a named-provider registry (`ctx.subagents`), unlike the single-implementation bash seam, so multiple transports (in-process, ACP, future A2A) coexist. This first PR lands the interface, a scripted test backend, and the model-facing tool, validated through the real cordis load path. - dsh-subagent: SubagentService registry + SubagentProvider/SubagentRun vocabulary + subagent/start|end events. Start-time capabilities (outputSchema, depthLimit, toolFilter) are checked pre-start and rejected loud; runtime capabilities (sendMessage, resume) are optional methods on SubagentRun. - dsh-subagent-mock (support): scripted provider for keyless, deterministic tests through the real Loader/export path. - dsh-tool-subagent: the model-facing `subagent` tool, config-bound to one provider; synchronous collect with try/finally dispose, signal->cancel bridging, and non-completed-stop-reason -> isError mapping. - Proposed RFC documenting the seam, the fork-vs-spawn-as-separate-backends decision, own-session isolation, synchronous-collect scope, and the deferral of background/poll/spill to a future unification with bash. - Wire the new group into tsconfigs, build refs, package hierarchy docs, the module graph, and the cordis catalog. RFC: docs/rfc/proposed/feature/2026-06-21-subagent-capability-seam.md
2026-06-21 22:31:56 +08:00
| `subagent` | `agent`, `llm`, `tools` |
refactor: order module-graph table topologically (low→high level) Replace the alphabetical package ordering with a Kahn-style topological sort (alphabetical tiebreak within each layer to stay deterministic), so the table and graph list leaf interfaces first and dependents last.
2026-06-16 21:11:09 +08:00
| `tool-bash` | `agent`, `bash`, `llm`, `tools` |
feat(tool-fs): result-time applied-hunk diffs for write/edit fs write/edit now emit a result-time contextual-diff tool_call_update (the applied hunk with ±3 context lines, one hunk per replace_all site), matching what claude-agent-acp sends and what makes an editor render the change in place. The call-time snippet diff stays; the result hunk supersedes it (ACP content-replace). Mechanism: - A persisted tool-private `meta` channel: execute may return `{ content, meta }`; `meta` (JsonValue) rides on the tool/result event and is handed back to presentResult, so the diff reproduces on replay (event-sourced). JsonValue is now exported from dsh-session. - The backend returns raw before/after text (storage facts) on FsWriteOutcome/FsEditOutcome; the tool computes the hunk via the npm `diff` package's structuredPatch. A create has no before → no result diff; a failed/aborted mutation carries no meta. - ToolResultView gains a DiffResultView; the bridge's result-side switch renders it as {type:'diff'} content blocks. RFC: docs/rfc/implemented/architecture/2026-07-02-result-time-applied-hunk-diffs.md (justifies the npm `diff` runtime dep over vendoring and the meta channel); the render-intent-union RFC's Non-goal is updated to record this shipped. All fs snapshot goldens re-recorded; edit/overwrite gain the contextual result diff, create/read/policy-reject unchanged in structure.
2026-07-03 17:12:00 +08:00
| `tool-fs` | `fs`, `llm`, `session`, `system-prompt`, `tools` |
feat(tool-todo): add the model-facing todo_write tool Add @deepseek-ai/dsh-tool-todo (a new packages/todo/ group): a model-facing todo_write(todos: [{content, status}]) tool with whole-list-replace semantics. Each call appends the full list as a todo/write event to the calling agent's session log; the current list is the most recent such event (last-write-wins). Single-owner — a non-agent caller is rejected. Beyond the schema's type/required/enum checks, execute rejects empty/duplicate content and more than one in_progress task, narrowing the loosely-typed args into a real TodoItem[]. Both UIs render off the existing session/event: the stdio UI prints a glyphed checklist; the ACP bridge maps the list to a `plan` sessionUpdate (todosToPlan synthesizes the priority ACP requires; status maps 1:1). Wired into the coding-agent, acp-agent, and snapshot example configs with a system-prompt nudge. Tests: unit (schema, validation, append/replace, no-agent rejection, presentCall, HMR-safety, Loader export-shape guard), full-loop integration through the agent loop, the ACP todosToPlan mapping + stream-update arm, the stdio render arm, and a session/load replay that re-emits the plan. New-group TS wiring added to tsconfig.base/json/build. RFC + a doc-inventory sweep (architecture, packages README, AGENTS layout, cookbook group list, example READMEs) ship with it. The todo-plan ACP snapshot scenario is recorded separately (needs an API key).
2026-06-29 10:30:52 +08:00
| `tool-todo` | `agent`, `session`, `tools` |
chore: adapt web seam to master's tsconfig + regenerate catalogs Register the five web packages in the root tsconfig.json project graph (master's typecheck moved to `tsc -b tsconfig.json` and dropped the separate tsconfig.typecheck.json), and regenerate the module graph and cordis catalog so they reflect the web packages on master.
2026-06-26 19:18:46 +08:00
| `tool-web` | `llm`, `system-prompt`, `tools`, `web` |
workflow: dynamic workflows — script-driven multi-agent orchestration A new capability family at packages/workflow/ in the bash seam shape, modeled on Claude Code's dynamic workflows: the model writes a JavaScript orchestration script (export const meta = {...} + plain-JS body), a runtime executes it, and the script — not the conversation — holds the loop, the branching, and the intermediate results. - dsh-workflow (ctx.workflows): abstract WorkflowService + run vocabulary (WorkflowRun whose result NEVER rejects) + observe-only workflow/* events carrying data snapshots (id + meta, never the live run), per-listener contained like subagent/*. - dsh-workflow-vm: in-process node:vm engine. Meta extraction via a string/comment-aware scanner (template interpolation rejected; literal evaluated alone in an empty timed context; statement blanked line- preservingly so stacks keep script line numbers). Hooks: agent(prompt, {label, phase, schema, model}) over ctx.subagents, parallel(), pipeline() (no cross-stage barrier), phase(), log(), args. Fatal-vs-null discipline: hook misuse (unknown/deferred options, bad arguments, unsupported schemas, tripped caps, seam start failures, cancellation) throws fatal WorkflowErrors the combinators RE-THROW — never dissolved into the per-item null reserved for child failures. Realm boundary: inbound values materialized by descriptor walks that never invoke accessors (defineProperty copies, __proto__-safe); outbound values rebuilt in-realm via the context's own JSON.parse. Determinism bans (Date.now/Math.random/argless new Date) kept so future resume support cannot break scripts. Caps and timeouts are validated Config. Every hook promise carries a no-op rejection consumer (app-boot exits on unhandled rejections). - dsh-tool-workflow: the model-facing workflow tool, synchronous like dsh-tool-subagent (start → await → try/finally dispose; abort bridged; non-completed → isError). Generic render card titled by a textual meta.name sniff. The tool description carries the authoring contract. Wired into examples/{coding-agent,acp-agent} with explicit-ask-only guidance. Coverage at every tier: unit (meta scanner, materializer incl. counting-getter and __proto__ regressions, combinator semantics, concurrency ceiling, caps, cancellation, no-unhandled-rejection abandon), integration over the real spawn stack, with-key e2e (real two-phase run + the tool through the registry pipeline), and a recorded ACP snapshot scenario (workflow-run, 1 child session). RFC: docs/rfc/implemented/feature/2026-07-05-dynamic-workflows.md (deferred work explicitly listed). AGENTS.md budget 1575 → 1590 for the new group's layout line.
2026-07-05 13:29:35 +08:00
| `tool-workflow` | `agent`, `llm`, `tools`, `workflow` |
refactor(examples): extract the app spine into dsh-agent-core + app packages Implements docs/rfc/.../2026-06-20-extract-example-app-packages.md. Each example was thick — a hand-rolled start.ts, an infra preamble, nested base.yml/base-core.yml/acp-tail.yml includes, and a coupled front-door cluster enforced only by prose. This moves the composition into packages so each example is a thin leaf cordis.yml: pick the swappable backends, load one app package. New packages: - @deepseek-ai/dsh-agent-core (packages/core/agent-core): one bundle plugin that loads the providerless/executor-less/UI-less spine (timer + llm + sessions + system-prompt + tools + agents + invariants + tool-bash + agent-loop) via ctx.plugin(...) inside apply(), and forwards agent-loop's `agents` list as its own Config (export const Config = AgentLoop.Config, default []). - @deepseek-ai/dsh-stdio-agent (packages/ui/stdio-agent): terminal chat APP — agent-core + console logger + readline UI + a pre-created `main` agent, with a bin. The demo:echo/coding front door. - @deepseek-ai/dsh-acp-agent (packages/ui/acp-agent): ACP server APP — agent-core + JSONL persistence + the acp bridge, NO stdout logger, with a bin. The stdout-purity footgun is structurally unreachable from the leaf. Amendment to the RFC: hmr stays a LEAF cordis.yml entry, not baked into dsh-stdio-agent. hmr is a Loader-only dev plugin (throws without --expose-internals; the in-process test tier can't even import its decorator form), so a package statically importing it could never carry the per-file coverage gate. Unlike the console logger, a stray hmr is not a stdout-purity footgun, so leaving it at the leaf costs no safety. With hmr out, all three new packages carry in-process unit specs at 100%. Boot glue (Loader tail, .env load, snapshot-mode selection, stdin-dispose lifecycle) moves into each app's bin; start.ts and base.yml/base-core.yml/ acp-tail.yml are deleted. Each app package gets a keyless real-load-path test that boots through its bin + the cordis Loader (guarding the unwrapExports export-shape bug class, postmortem 0001). ACP snapshot replay stays green against the existing committed goldens (pure boot restructuring). RFC moved proposed->implemented with the amendment recorded; package/example/architecture docs and the module graph updated.
2026-06-21 12:03:44 +08:00
| `agent-core` | `agent`, `agent-loop`, `invariants`, `llm`, `session`, `system-prompt`, `tool-bash`, `tools` |
feat(hooks): dsh-hooks-claude + dsh-hooks-codex bridges (hooks stack PR-F) The two bridge plugins that run a user's existing Claude Code / Codex hook config on the harness's typed interception seams, built on the shared dsh-hook-protocol library. A bridge is a faithfulness adapter, not a power tool: anything it does a native cordis plugin does more powerfully — the bridge exists only to run UNMODIFIED external hooks. - dsh-hooks-claude: CC dialect. Seven hook points (SessionStart, UserPromptSubmit, PreToolUse, PostToolUse, Stop, SubagentStart, SubagentStop), CC per-event stdin payloads, env + ${CLAUDE_PLUGIN_ROOT}/ ${CLAUDE_PROJECT_DIR} substitution, literal-or-regex matcher. - dsh-hooks-codex: Codex dialect — a deliberate subset. Five hook points, always-regex matcher, snake_case payloads (turn_id/model, no trailing newline), no env/substitution, block-only decisions. Both map the neutral merged outcome onto the seam's typed Decision and stamp an explicit {kind:'plugin'} source on injected context (so it is never mislabeled as a user prompt). Config parse-failure is contained; only command hooks run. updatedInput is logged+warned (input rewrite deferred); the Stop loop-guard is deferred (TODO). Tests: per-file 100% — config-parse unit branches + per-seam mappings end-to-end through the REAL loop + REAL bash + REAL shell scripts (scripted mock model only) + a real-Loader export-shape guard. A keyless ACP snapshot scenario (hook-prompt-block) proves a UserPromptSubmit hook blocks a prompt end-to-end (rejected turn -> ACP cancelled, hook/* events in the log); a with-key e2e (hooks.e2e.ts) proves a PreToolUse hook blocks real bash (verified on disk). The snapshot normalizer now scrubs hook/result.durationMs. RFC: docs/rfc/implemented/feature/2026-06-30-hook-bridges.md
2026-07-01 04:22:00 +08:00
| `hooks-claude` | `agent`, `hook-protocol`, `llm`, `session`, `subagent`, `tools` |
Add the ACP subagent backend: out-of-process delegation (PR3) The first OUT-OF-PROCESS subagent backend, proving the seam generalizes past the in-process backends. @deepseek-ai/dsh-subagent-acp runs each child agent in a spawned subprocess, driven over the Agent Client Protocol as the CLIENT — the direction-inverted twin of the dsh-acp server bridge. Point the configured command at the acp-agent example and the harness talks to its own process. - Fresh process per run: start spawns, runs one ACP session (initialize → newSession → prompt), dispose kills the subprocess and awaits its exit. - Minimal client stub: advertises no fs/terminal; accumulates agent_message_chunk text as the result output; auto-answers session/request_permission by a configured policy (reject default / allow). No start-time capabilities (an out-of-process child can't enforce the parent's depth/tool-filter); ignores request.parent; injects only `subagents`. - StopReason mapping (end_turn→completed, cancelled→aborted, …); result resolves error/aborted on a child failure, never rejects (seam contract). - Security: credential-shaped ambient env vars are scrubbed; the child's own key is forwarded only via explicit config.env. A spawn-level error (ENOENT) is captured and raced against the ACP drive so a bad command settles error rather than crashing the parent. Testing designed at every tier: keyless integration drives a scripted mock ACP server subprocess (cancellation incl. the pre-newSession race and a torn-pipe-after-cancel, permission auto-answer, non-message updates, spawn failure, HMR, export shape) at 100% coverage; a with-key e2e drives the REAL acp-agent example process (PONG + real file write, verified on disk) — the harness driving itself. Snapshot coverage of an ACP child is deferred as TODO(acp-subagent-replay) (each child is its own process with its own replay). Stayed on @agentclientprotocol/sdk 0.25.1: the proposed 0.28.x bump only deprecates the stable ClientSideConnection/AgentSideConnection API this layer uses (33 sites incl. the server bridge), turning no-deprecated red across code this PR shouldn't rewrite — that fluent-API migration is its own follow-up. The backend needs nothing 0.28.x adds. This completes the subagent seam stack (PR1 interface → PR2 in-process → PR2.5 snapshot infra → PR3 ACP); the seam RFC moves to implemented/, amended.
2026-06-22 10:47:02 +08:00
| `subagent-acp` | `agent`, `llm`, `subagent` |
subagent: implement structured output for in-process backends The seam vocabulary (SubagentStartRequest.outputSchema, SubagentResult .structured) existed but no in-process backend honored it — spawn/fork advertised outputSchema: false. This lands the missing half: - dsh-tools gains a structured-output JSON Schema subset (json-schema.ts): StructuredOutputSchema, assertSupportedOutputSchema (rejects loud outside the enforced subset, every violation listed), validateStructuredValue (path-qualified issues, total). outputSchema's seam type becomes this raw JSON-Schema subset instead of the author-facing SchemaSpec DSL — the schema travels verbatim to the model as a forced tool's parameters. - dsh-subagent-inprocess gains the shared structured runtime: one global structured_output capture tool (placeholder parameters) + a prepend:true agent/request listener doing FINAL-REQUEST enforcement (strip for plain agents, per-run schema for structured children — survives downstream request-replacing listeners) + an agent/turn-continuation veto that stops a child's turn once captured (no wasted extra model step). Lifetime is refcounted by backends (plugin lifetime) AND live runs (start→settle). - startInProcessRun drives the capture: subset asserted before the child exists, instruction appended to the child's system prompt, clean-finish nudge loop (structuredNudgeRetries, backend Config, default 1), captured value on result.structured; a clean finish without a capture settles 'error' (never a silent success with a missing field). - spawn + fork flip outputSchema: true and inject 'tools'.
2026-07-05 11:35:39 +08:00
| `subagent-inprocess` | `agent`, `llm`, `session`, `subagent`, `tools` |
Add subagent capability seam: interface, mock backend, model-facing tool Introduce the `packages/subagent/` group and the abstract subagent seam — an agent delegating to a child agent — as a named-provider registry (`ctx.subagents`), unlike the single-implementation bash seam, so multiple transports (in-process, ACP, future A2A) coexist. This first PR lands the interface, a scripted test backend, and the model-facing tool, validated through the real cordis load path. - dsh-subagent: SubagentService registry + SubagentProvider/SubagentRun vocabulary + subagent/start|end events. Start-time capabilities (outputSchema, depthLimit, toolFilter) are checked pre-start and rejected loud; runtime capabilities (sendMessage, resume) are optional methods on SubagentRun. - dsh-subagent-mock (support): scripted provider for keyless, deterministic tests through the real Loader/export path. - dsh-tool-subagent: the model-facing `subagent` tool, config-bound to one provider; synchronous collect with try/finally dispose, signal->cancel bridging, and non-completed-stop-reason -> isError mapping. - Proposed RFC documenting the seam, the fork-vs-spawn-as-separate-backends decision, own-session isolation, synchronous-collect scope, and the deferral of background/poll/spill to a future unification with bash. - Wire the new group into tsconfigs, build refs, package hierarchy docs, the module graph, and the cordis catalog. RFC: docs/rfc/proposed/feature/2026-06-21-subagent-capability-seam.md
2026-06-21 22:31:56 +08:00
| `subagent-mock` | `agent`, `llm`, `subagent` |
| `tool-subagent` | `agent`, `llm`, `subagent`, `tools` |
workflow: dynamic workflows — script-driven multi-agent orchestration A new capability family at packages/workflow/ in the bash seam shape, modeled on Claude Code's dynamic workflows: the model writes a JavaScript orchestration script (export const meta = {...} + plain-JS body), a runtime executes it, and the script — not the conversation — holds the loop, the branching, and the intermediate results. - dsh-workflow (ctx.workflows): abstract WorkflowService + run vocabulary (WorkflowRun whose result NEVER rejects) + observe-only workflow/* events carrying data snapshots (id + meta, never the live run), per-listener contained like subagent/*. - dsh-workflow-vm: in-process node:vm engine. Meta extraction via a string/comment-aware scanner (template interpolation rejected; literal evaluated alone in an empty timed context; statement blanked line- preservingly so stacks keep script line numbers). Hooks: agent(prompt, {label, phase, schema, model}) over ctx.subagents, parallel(), pipeline() (no cross-stage barrier), phase(), log(), args. Fatal-vs-null discipline: hook misuse (unknown/deferred options, bad arguments, unsupported schemas, tripped caps, seam start failures, cancellation) throws fatal WorkflowErrors the combinators RE-THROW — never dissolved into the per-item null reserved for child failures. Realm boundary: inbound values materialized by descriptor walks that never invoke accessors (defineProperty copies, __proto__-safe); outbound values rebuilt in-realm via the context's own JSON.parse. Determinism bans (Date.now/Math.random/argless new Date) kept so future resume support cannot break scripts. Caps and timeouts are validated Config. Every hook promise carries a no-op rejection consumer (app-boot exits on unhandled rejections). - dsh-tool-workflow: the model-facing workflow tool, synchronous like dsh-tool-subagent (start → await → try/finally dispose; abort bridged; non-completed → isError). Generic render card titled by a textual meta.name sniff. The tool description carries the authoring contract. Wired into examples/{coding-agent,acp-agent} with explicit-ask-only guidance. Coverage at every tier: unit (meta scanner, materializer incl. counting-getter and __proto__ regressions, combinator semantics, concurrency ceiling, caps, cancellation, no-unhandled-rejection abandon), integration over the real spawn stack, with-key e2e (real two-phase run + the tool through the registry pipeline), and a recorded ACP snapshot scenario (workflow-run, 1 child session). RFC: docs/rfc/implemented/feature/2026-07-05-dynamic-workflows.md (deferred work explicitly listed). AGENTS.md budget 1575 → 1590 for the new group's layout line.
2026-07-05 13:29:35 +08:00
| `workflow-vm` | `agent`, `brand`, `llm`, `subagent`, `tools`, `workflow` |
feat(ui): share the app bins' boot glue in @deepseek-ai/dsh-app-boot The four near-twin helpers the two published bins carried — loadEnv, installFailLoud, assertEntriesLoaded, boot — live once in packages/ui/app-boot, parameterized by the bin's diagnostic prefix and injectable at their side-effect seams (warn sink, process slice), so every branch sits under the per-file 100% coverage gate: the unit suite drives boot() in-process against the real Loader (relative-specifier configs) through both the settled-tree path and the fiber-less-entry rejection, and exercises the ENOENT/unloadable .env split, the Error/non-Error/stackless fail-loud arms, and the disabled-entry exclusion. resolveConfigPath (snapshot-aware) becomes the single path resolver for both bins. Each bin.ts is now a thin self-executing composition plus its app-specific lifecycle (acp: replay env-skip + stdin-EOF dispose; stdio: nothing extra), exports nothing, and stays coverage-excluded; the built-bin smokes still prove both artifacts under plain node in the node_modules-shaped temp dir (now symlinking ui/app-boot), including the missing-config non-zero exit. Implements docs/rfc/implemented/simplification/2026-07-04-share-app-bin-boot-glue.md (moved from proposed/ and amended); the extract-example-app-packages RFC's bin-ownership facts are amended in the same change.
2026-07-04 16:43:06 +08:00
| `acp-agent` | `acp`, `agent-core`, `app-boot`, `session-persistence-jsonl` |
Merge remote-tracking branch 'origin/simpl-b1-fold-ui-stdio' into simpl-b2-app-boot # Conflicts: # docs/module-graph.md # packages/README.md # pnpm-lock.yaml
2026-07-04 17:20:43 +08:00
| `stdio-agent` | `agent`, `agent-core`, `app-boot`, `llm`, `session`, `session-persistence-jsonl` |
Extract the shared in-process driver into dsh-subagent-inprocess (review feedback) The shared run driver lived inside dsh-subagent-spawn, so the spawn package carried fork-aware seeding logic and dsh-subagent-fork depended backward on dsh-subagent-spawn — the two in-process backends were not independent. Move the driver (startInProcessRun, depthOf, SubagentDepthError, InProcessRunOptions) into a new pure-library package @deepseek-ai/dsh-subagent-inprocess that registers nothing. spawn and fork now both depend only on that driver and neither knows about the other; spawn no longer re-exports it and fork no longer imports from spawn. Also wire BOTH backends in examples/coding-agent/cordis.yml (config-only): load dsh-subagent-spawn + dsh-subagent-fork + two dsh-tool-subagent instances with distinct toolNames (subagent → spawn, subagent_fork → fork), demonstrating that exposing multiple transports needs no code change.
2026-06-22 14:27:38 +08:00
| `subagent-fork` | `agent`, `session`, `subagent`, `subagent-inprocess` |
| `subagent-spawn` | `subagent`, `subagent-inprocess` |
Reference in a new issue Copy permalink