kevchess15/deepseek-harness
1
1
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions 1
deepseek-harness/.agents/notes/implemented/architecture/2026-06-11-content-block-vocabulary.md

29 lines
2.6 KiB
Markdown
Raw Normal View History

Rename RFCs to Agent Notes
2026-07-19 22:50:49 +08:00
# Agent Note: Provider-neutral content-block vocabulary owned by dsh-llm
Backfill architecture decision records Seven ADRs capturing the why behind decisions already made: vendoring Cordis as source with a guarded manifest; the microkernel event taxonomy with one swappable concrete loop; event-sourced sessions with derived history and the append-before-emit ordering contract; the provider-neutral content-block vocabulary (and why not OpenAI/Anthropic shapes); the custom tool-schema DSL over schemastery; tool schemas living in the prompt assembly; and mechanical quality gates over prose guidelines (the agents-write-the-code rationale).
2026-06-11 15:24:14 +08:00
docs(rfc): define and enforce a uniform RFC format; adopt it across the corpus Define the in-file RFC contract in docs/rfc/README.md § The file format: the header block (`# RFC: <title>` plus a dateless Status enum cross-checked against the lifecycle folder), the per-lifecycle body skeleton (a Problem opener everywhere; Proposal/Alternatives considered/ Acceptance criteria/Risks in proposed/; present-tense Decision/ Consequences with proposal-era headings banned in implemented/; the frozen proposal shape in rejected/), and a mandatory Alternatives considered section with a date-fenced grandfather comment for pre-format RFCs whose alternatives are not reconstructible from the record. Enforce it with a new doc-sync gate, scripts/verify-rfc-format.ts, and normalize all 112 RFCs to it: ~15 Status-line spellings collapse to the enum, 29 Context openers become Problem, the 39 legacy-format XXX debt markers are resolved and banned from reappearing, proposal-era sections in implemented RFCs are rewritten to shipped reality (including the web/fs/subagent seam RFCs' migration plans and test checklists, closing the doc-tiers deferred-work item on the web seam), every RFC gains an Alternatives considered section or the grandfather comment, and the bilingual pair is re-mirrored and re-recorded. Move the generated index tables out of README.md into a fully generated docs/rfc/INDEX.md — gen-rfc-index now writes the whole file, and verify-rfc-classification checks its freshness and rejects index-shaped rows in the curated README — which makes room for the format contract to live in the README front door instead of a separate FORMAT.md. The decision record, and the first RFC written in the new format, is docs/rfc/implemented/process/2026-07-05-uniform-rfc-format.md.
2026-07-05 22:58:25 +08:00
Status: implemented
docs: unify ADR/RFC trees into one lifecycle-organized RFC tree Collapse docs/adr/ and docs/rfc/ into a single docs/rfc/ with proposed/, implemented/, and rejected/ subfolders. Every file is renamed to yyyy-mm-dd-topic-title.md, where the date is when the topic was first proposed (from git history). ADRs and RFCs that covered exactly the same topic are merged (property-based testing, session persistence); the umbrella RFC 005 stays split across its three implemented decisions, and RFC 006's deferred part-3 (API extractor reports) splits into its own proposed RFC. All cross-references become machine-checkable relative links instead of bare "ADR NNNN" / "RFC NNN" prose. Add a verify-md-links doc-sync gate (scripts/verify-md-links.ts) that checks every relative Markdown cross-link resolves, wired into doc-sync alongside verify-md-wrap. This makes the reorganization self-verifying: the same change that rewrote ~forty inter-doc links adds the check that proves none dangle. Document the cross-link convention in a new docs/AGENTS.md and record the gate as an implemented RFC. doc-sync, typecheck, lint, and the full test suite (667) all pass.
2026-06-18 02:18:24 +08:00
docs(i18n): address RFC translation review
2026-07-22 22:29:39 +08:00
English | [中文](2026-06-11-content-block-vocabulary.zh.md)
docs(rfc): define and enforce a uniform RFC format; adopt it across the corpus Define the in-file RFC contract in docs/rfc/README.md § The file format: the header block (`# RFC: <title>` plus a dateless Status enum cross-checked against the lifecycle folder), the per-lifecycle body skeleton (a Problem opener everywhere; Proposal/Alternatives considered/ Acceptance criteria/Risks in proposed/; present-tense Decision/ Consequences with proposal-era headings banned in implemented/; the frozen proposal shape in rejected/), and a mandatory Alternatives considered section with a date-fenced grandfather comment for pre-format RFCs whose alternatives are not reconstructible from the record. Enforce it with a new doc-sync gate, scripts/verify-rfc-format.ts, and normalize all 112 RFCs to it: ~15 Status-line spellings collapse to the enum, 29 Context openers become Problem, the 39 legacy-format XXX debt markers are resolved and banned from reappearing, proposal-era sections in implemented RFCs are rewritten to shipped reality (including the web/fs/subagent seam RFCs' migration plans and test checklists, closing the doc-tiers deferred-work item on the web seam), every RFC gains an Alternatives considered section or the grandfather comment, and the bilingual pair is re-mirrored and re-recorded. Move the generated index tables out of README.md into a fully generated docs/rfc/INDEX.md — gen-rfc-index now writes the whole file, and verify-rfc-classification checks its freshness and rejects index-shaped rows in the curated README — which makes room for the format contract to live in the README front door instead of a separate FORMAT.md. The decision record, and the first RFC written in the new format, is docs/rfc/implemented/process/2026-07-05-uniform-rfc-format.md.
2026-07-05 22:58:25 +08:00
## Problem
Backfill architecture decision records Seven ADRs capturing the why behind decisions already made: vendoring Cordis as source with a guarded manifest; the microkernel event taxonomy with one swappable concrete loop; event-sourced sessions with derived history and the append-before-emit ordering contract; the provider-neutral content-block vocabulary (and why not OpenAI/Anthropic shapes); the custom tool-schema DSL over schemastery; tool schemas living in the prompt assembly; and mechanical quality gates over prose guidelines (the agents-write-the-code rationale).
2026-06-11 15:24:14 +08:00
docs(rfc): define and enforce a uniform RFC format; adopt it across the corpus Define the in-file RFC contract in docs/rfc/README.md § The file format: the header block (`# RFC: <title>` plus a dateless Status enum cross-checked against the lifecycle folder), the per-lifecycle body skeleton (a Problem opener everywhere; Proposal/Alternatives considered/ Acceptance criteria/Risks in proposed/; present-tense Decision/ Consequences with proposal-era headings banned in implemented/; the frozen proposal shape in rejected/), and a mandatory Alternatives considered section with a date-fenced grandfather comment for pre-format RFCs whose alternatives are not reconstructible from the record. Enforce it with a new doc-sync gate, scripts/verify-rfc-format.ts, and normalize all 112 RFCs to it: ~15 Status-line spellings collapse to the enum, 29 Context openers become Problem, the 39 legacy-format XXX debt markers are resolved and banned from reappearing, proposal-era sections in implemented RFCs are rewritten to shipped reality (including the web/fs/subagent seam RFCs' migration plans and test checklists, closing the doc-tiers deferred-work item on the web seam), every RFC gains an Alternatives considered section or the grandfather comment, and the bilingual pair is re-mirrored and re-recorded. Move the generated index tables out of README.md into a fully generated docs/rfc/INDEX.md — gen-rfc-index now writes the whole file, and verify-rfc-classification checks its freshness and rejects index-shaped rows in the curated README — which makes room for the format contract to live in the README front door instead of a separate FORMAT.md. The decision record, and the first RFC written in the new format, is docs/rfc/implemented/process/2026-07-05-uniform-rfc-format.md.
2026-07-05 22:58:25 +08:00
The harness needs one internal language for messages that the loop, session log, and all plugins speak.
Backfill architecture decision records Seven ADRs capturing the why behind decisions already made: vendoring Cordis as source with a guarded manifest; the microkernel event taxonomy with one swappable concrete loop; event-sourced sessions with derived history and the append-before-emit ordering contract; the provider-neutral content-block vocabulary (and why not OpenAI/Anthropic shapes); the custom tool-schema DSL over schemastery; tool schemas living in the prompt assembly; and mechanical quality gates over prose guidelines (the agents-write-the-code rationale).
2026-06-11 15:24:14 +08:00
## Decision
docs(rfc): define and enforce a uniform RFC format; adopt it across the corpus Define the in-file RFC contract in docs/rfc/README.md § The file format: the header block (`# RFC: <title>` plus a dateless Status enum cross-checked against the lifecycle folder), the per-lifecycle body skeleton (a Problem opener everywhere; Proposal/Alternatives considered/ Acceptance criteria/Risks in proposed/; present-tense Decision/ Consequences with proposal-era headings banned in implemented/; the frozen proposal shape in rejected/), and a mandatory Alternatives considered section with a date-fenced grandfather comment for pre-format RFCs whose alternatives are not reconstructible from the record. Enforce it with a new doc-sync gate, scripts/verify-rfc-format.ts, and normalize all 112 RFCs to it: ~15 Status-line spellings collapse to the enum, 29 Context openers become Problem, the 39 legacy-format XXX debt markers are resolved and banned from reappearing, proposal-era sections in implemented RFCs are rewritten to shipped reality (including the web/fs/subagent seam RFCs' migration plans and test checklists, closing the doc-tiers deferred-work item on the web seam), every RFC gains an Alternatives considered section or the grandfather comment, and the bilingual pair is re-mirrored and re-recorded. Move the generated index tables out of README.md into a fully generated docs/rfc/INDEX.md — gen-rfc-index now writes the whole file, and verify-rfc-classification checks its freshness and rejects index-shaped rows in the curated README — which makes room for the format contract to live in the README front door instead of a separate FORMAT.md. The decision record, and the first RFC written in the new format, is docs/rfc/implemented/process/2026-07-05-uniform-rfc-format.md.
2026-07-05 22:58:25 +08:00
Own the vocabulary: messages are arrays of typed content blocks (`text`, `reasoning`, `tool-call`, `tool-result`), with the union derived from the merge-extensible `ContentBlockMap` so plugins add block types via declaration merging. The same merge-extensible-map pattern types every "stringly" field (`MessageSource`, `FinishReason`, `TurnTrigger`, `TurnEndReason`). Streaming is a raw chunk protocol; `BlockAssembler` is the single shared assembly implementation. Adapters translate to provider wire formats — mapping cost lives in adapters, where it belongs.
Backfill architecture decision records Seven ADRs capturing the why behind decisions already made: vendoring Cordis as source with a guarded manifest; the microkernel event taxonomy with one swappable concrete loop; event-sourced sessions with derived history and the append-before-emit ordering contract; the provider-neutral content-block vocabulary (and why not OpenAI/Anthropic shapes); the custom tool-schema DSL over schemastery; tool schemas living in the prompt assembly; and mechanical quality gates over prose guidelines (the agents-write-the-code rationale).
2026-06-11 15:24:14 +08:00
refactor(agent-loop): drop the steering/message session event Steer, inject, and followup now land as durable user/message events on the session surface; the steering/message event type and its ConversationNode kind are removed from the client projection. Update tests, docs, generated catalogs, and agent notes to match, and align the steering e2e fixture and prompt inventory assertions with the durable user/message landing.
2026-08-03 16:04:26 +08:00
In-session context injection (`context/message`) and mid-turn steering originally rendered as tagged user-role envelopes (the system-reminder pattern) rather than a new role, so adapters carry zero burden. Both now project as plain user content with no wrapper; see [the injected-content-envelope Agent Note](../simplification/2026-07-20-unwrap-injected-content-envelopes.md). Live-adapter validation confirms this rendering for current DeepSeek behavior; a future provider-specific mismatch belongs in that adapter rather than a new canonical role.
Backfill architecture decision records Seven ADRs capturing the why behind decisions already made: vendoring Cordis as source with a guarded manifest; the microkernel event taxonomy with one swappable concrete loop; event-sourced sessions with derived history and the append-before-emit ordering contract; the provider-neutral content-block vocabulary (and why not OpenAI/Anthropic shapes); the custom tool-schema DSL over schemastery; tool schemas living in the prompt assembly; and mechanical quality gates over prose guidelines (the agents-write-the-code rationale).
2026-06-11 15:24:14 +08:00
docs(rfc): define and enforce a uniform RFC format; adopt it across the corpus Define the in-file RFC contract in docs/rfc/README.md § The file format: the header block (`# RFC: <title>` plus a dateless Status enum cross-checked against the lifecycle folder), the per-lifecycle body skeleton (a Problem opener everywhere; Proposal/Alternatives considered/ Acceptance criteria/Risks in proposed/; present-tense Decision/ Consequences with proposal-era headings banned in implemented/; the frozen proposal shape in rejected/), and a mandatory Alternatives considered section with a date-fenced grandfather comment for pre-format RFCs whose alternatives are not reconstructible from the record. Enforce it with a new doc-sync gate, scripts/verify-rfc-format.ts, and normalize all 112 RFCs to it: ~15 Status-line spellings collapse to the enum, 29 Context openers become Problem, the 39 legacy-format XXX debt markers are resolved and banned from reappearing, proposal-era sections in implemented RFCs are rewritten to shipped reality (including the web/fs/subagent seam RFCs' migration plans and test checklists, closing the doc-tiers deferred-work item on the web seam), every RFC gains an Alternatives considered section or the grandfather comment, and the bilingual pair is re-mirrored and re-recorded. Move the generated index tables out of README.md into a fully generated docs/rfc/INDEX.md — gen-rfc-index now writes the whole file, and verify-rfc-classification checks its freshness and rejects index-shaped rows in the curated README — which makes room for the format contract to live in the README front door instead of a separate FORMAT.md. The decision record, and the first RFC written in the new format, is docs/rfc/implemented/process/2026-07-05-uniform-rfc-format.md.
2026-07-05 22:58:25 +08:00
## Alternatives considered
- **Mirror the DeepSeek/OpenAI chat-completions shape** — zero mapping cost for the first provider, but awkward for rich content (reasoning, tool results as structured blocks).
- **Adopt Anthropic's Messages block structure verbatim** — battle-tested, but the canonical types would mirror a third-party API the harness does not target first.
Backfill architecture decision records Seven ADRs capturing the why behind decisions already made: vendoring Cordis as source with a guarded manifest; the microkernel event taxonomy with one swappable concrete loop; event-sourced sessions with derived history and the append-before-emit ordering contract; the provider-neutral content-block vocabulary (and why not OpenAI/Anthropic shapes); the custom tool-schema DSL over schemastery; tool schemas living in the prompt assembly; and mechanical quality gates over prose guidelines (the agents-write-the-code rationale).
2026-06-11 15:24:14 +08:00
## Consequences
docs: tighten prose audit after master retarget
2026-07-13 16:24:32 +08:00
- Reasoning has a core home without provider-specific shapes.
feat(llm-deepseek): support multimodal requests
2026-08-19 11:39:41 +08:00
- Multimodal blocks return only with coordinated adapter, UI, and compaction support; see [the drop-image Agent Note](../../archived/simplification/2026-07-04-drop-image-content-block.md).
docs(notes): archive low-value decision records
2026-07-26 23:06:00 +08:00
- Cache hints and assistant prefill remain absent until a shipping adapter can honor them; see the [producer-less variants](../../archived/simplification/2026-07-04-prune-producerless-vocabulary-variants.md) and [inert request knobs](../../archived/simplification/2026-07-04-drop-inert-request-knobs.md) Agent Notes.
docs: retire completed tagged-envelope review TODO
2026-06-19 09:28:23 +08:00
- Every adapter pays a translation cost; the first real adapters have since validated the streaming protocol, and new adapters should continue proving their provider-specific mapping in adapter-local tests.
docs(notes): rename CallId to ToolCallId in agent notes Keep the active implemented and proposed Agent Notes current with the renamed ToolCallId brand: branded-ids, content-block-vocabulary, approval-seam, tool-output-spill-files, and task-surface, in both languages, with re-recorded .i18n.yaml pairing records.
2026-08-19 22:35:31 +08:00
- IDs that cross package boundaries are branded (`ToolCallId`, the shared agent/session `SessionId`) — nominal typing at zero runtime cost.
Reference in a new issue Copy permalink