deepseek-harness/docs/subsystems/tasks.md

290 lines
14 KiB
Markdown
Raw Normal View History

# Background Task Runtime
English | [中文](tasks.zh.md)
2026-08-09 15:27:21 +08:00
Types shared by long-running producers, `ctx.tasks`, and task controls. The [runtime Agent Note](../../.agents/notes/implemented/architecture/2026-06-20-generic-long-running-tool-runtime.md) owns the design; this page records the exact fields and variants from [`packages/tasks/tasks/src/types.ts`](../../packages/tasks/tasks/src/types.ts).
## Ids and status
`TaskId` is a [branded id](core.md#branded-ids) generated as `<kind>-N`. Access control relies on owner authorization, not id secrecy. `TaskKind` derives from a merge-extensible map; the registry treats kinds as opaque id namespaces.
```ts type-equiv
/**
* Producer-defined task kinds. Plugins extend this map by declaration merging;
* the registry treats every value as an opaque id namespace.
*/
interface TaskKindMap {
bash: 'bash'
subagent: 'subagent'
}
```
`TaskStatus` is `'running' | 'stopping' | 'completed' | 'killed' | 'failed'`; producer-specific facts belong in `TaskSnapshot.detail`.
## Producer contract
`TaskStart` declares identity and a starter. The runtime finishes preflight before calling `run()` and commits without a later failable step. Producers own execution resources; the runtime owns identity, access, and lifecycle state.
```ts type-equiv
/**
* Producer declaration passed to {@link TaskService.start}. The runtime
* preflights access and cleanup before invoking {@link run}; the producer owns
* execution resources while the runtime owns identity and lifecycle state.
*/
interface TaskStart {
/** Producer kind — also the id prefix (`bash`, `subagent`, …). */
kind: TaskKind
/** One-line model-facing label (the command; the delegation description). */
label: string
2026-07-22 22:37:20 +08:00
/**
* Optional UTF-8 byte cap for each complete model-facing completion notice or
* output read, including controller status metadata.
2026-07-22 22:37:20 +08:00
*/
outputLimitBytes?: number
/**
* Owning live agent. Access is fenced by its session id, and agent disposal
* cancels and awaits the task. The instance must be the one currently
* registered under its agent id. Omitting the owner creates an unowned task,
* open to any caller until service disposal.
*/
owner?: Agent
/**
* Start the work after preflight and synchronously return its hooks. Called
* once; a throw leaves nothing registered, and the producer must clean up any
* partially started resources.
*/
run(): TaskHooks
}
```
2026-08-09 15:27:21 +08:00
`TaskHooks.done` resolves after the producer releases its resources, not merely when work finishes. Optional `readOutput` distinguishes consuming stream tasks from final-output-only tasks.
```ts type-equiv
/** Hooks through which the runtime controls and observes producer work. */
interface TaskHooks {
/**
* Request termination. Must be synchronous, idempotent, and eventually settle
* {@link done}; throws propagate. The optional reason is forwarded verbatim.
*/
cancel(reason?: string): void
/**
* Resolves after the producer releases its resources, not merely when work
* finishes. Must not reject; the runtime converts a rejection to `failed`.
* If teardown cancellation throws, the runtime may force-fail only the
* registry record without claiming that the work stopped.
*/
done: Promise<TaskOutcome>
/**
* Consume output produced since the previous call. The producer formats
* truncation and spill notices. Absence marks a final-output-only task; each
* task has one consuming cursor.
*/
readOutput?(): string
}
```
```ts type-equiv
/** Terminal result supplied by a producer through {@link TaskHooks.done}. */
interface TaskOutcome {
/** How the task ended: finished (`completed`), cancelled (`killed`), or broke (`failed`). */
status: 'completed' | 'killed' | 'failed'
/** Kind-specific detail rendered into status lines ('exit code: 3', 'max-tokens'). */
detail?: string
/** Final output for tasks without `readOutput`; stream tasks leave it unset. */
output?: string
}
```
## Consumer views
Snapshots are fresh read-only projections. `ownerSession` carries the shared `SessionId` used for authorization; completion listeners separately receive the exact owner object used for lifecycle cleanup. `reported` suppresses a completion notice after another reporter has delivered or committed to deliver the terminal state, including the teardown cancel that drains an owner or the service.
```ts type-equiv
/**
* A read-only projection of one task, safe to hand to listeners and tools —
* a fresh object per call, never live registry state.
*/
interface TaskSnapshot {
/** The registry-issued id (`<kind>-N`). */
id: TaskId
/** The producer kind the task was registered with. */
kind: TaskKind
/** The producer-supplied one-line label. */
label: string
2026-07-22 22:37:20 +08:00
/** Producer-owned cap for complete model-facing notices and output reads. */
outputLimitBytes?: number
/**
* Owner session id used for authorization and correlation; absent for
* unowned tasks. Completion listeners receive the exact {@link Agent}
* separately through {@link TaskDoneListener}.
*/
ownerSession?: SessionId
/** Current lifecycle state. */
status: TaskStatus
/** Kind-specific status detail, present once the producer supplied one (usually terminal). */
detail?: string
/** Epoch ms when the task was registered. */
startedAt: number
/** Epoch ms when the task settled; absent while `running`/`stopping`. */
finishedAt?: number
/**
* True when a kill, read, wait, or teardown cancel has reported or committed
* to report the terminal state. Completion reporters suppress redundant
* notices when set. Teardown claims it because the owner or service being
* destroyed leaves no reader: a reporter that opens a turn on notice would
* otherwise spend a model request per teardown layer.
*/
reported: boolean
}
```
```ts type-equiv
/** Output and post-read state returned by {@link TaskService.read}. */
interface TaskRead {
/**
* Stream kinds: the consuming delta since the previous read. Final-output
* kinds: empty while live, the terminal {@link TaskOutcome.output} (or
* empty) once settled — idempotent, never consumed.
*/
text: string
/** The task's state at read time. */
snapshot: TaskSnapshot
}
```
## Service behavior
The abstract [`TaskService`](../../packages/tasks/tasks/src/index.ts) Service Definition specifies atomic `start`, caller-scoped `get` and `list`, `read`, `kill`, bounded `wait`, failure-isolated `onTaskDone` and `onTasksChanged` listeners, and when `attachController` becomes available; [`LocalTaskService`](../../packages/tasks/tasks-local/src/index.ts) is the process-local Service provider. Authorization compares owner sessions; owner cleanup selects the exact registered `Agent` instance. See [`dsh-tasks`](../../packages/tasks/tasks/README.md) for the Service Definition contract, [`dsh-tasks-local`](../../packages/tasks/tasks-local/README.md) for the registry lifecycle, and [`dsh-tool-tasks`](../../packages/tasks/tool-tasks/README.md) for the model-facing Consumer.
<!-- BEGIN GENERATED cordis-surface (gen-cordis-catalog.ts) — do not edit between markers -->
<a id="cordis-surface"></a>
## Cordis API
Generated from source by `scripts/gen-cordis-catalog.ts` (verified fresh by `pnpm run verify-cordis-catalog` in doc-sync; regenerate with `pnpm run gen-cordis-catalog`) — this section is byte-identical in both language sides of the page. Signature blocks use a `ts cordis-catalog` fence and keep the original source JSDoc; dispatch modes are defined in the [primer](../cordis-primer.md#dispatch-modes), and the framework-inherited `ctx` API lives in [cordis-api/inherited.md](../cordis-api/inherited.md).
<a id="ctxtasks--taskservice-abstract-seam"></a>
### `ctx.tasks` — `TaskService` (abstract seam)
Abstract background task registry. Subclass, implement the abstract methods, and load the subclass as a plugin — it registers as `ctx.tasks` (one implementation per context; loading a second throws, which is cordis' standard duplicate-service behavior).
Implementations must honor these semantics:
- Registrations outlive producer and controller fibers. Owner and service disposal cancel live work and await compliant producers; a throwing teardown cancel force-fails only the record. Teardown cancellation also marks the record reported, because a record its owner is being destroyed for has no reader left.
- Owned-task access is fenced by the owner's session id. Ids are predictable, so authorization — not secrecy — is the boundary.
- Settlement is first-wins: one terminal record, released waiters, and one round of contained listener notification, even against a late producer outcome. Completion is announced last, after the record is committed and every other observer of the settlement has seen it, because a reporter may open a model turn synchronously.
- start refuses work while no attached task controller serves the spec's owner, so a producer cannot start work that owner cannot collect or stop. One registry serves every composition in the process, so this question — and completion-listener delivery — is owner-relative rather than process-wide: registrations made from an unscoped context serve every owner, and registrations made under an agent composition's scope serve exactly the agents composed under it.
```ts cordis-catalog
/**
* Preflight access, validation, and owner cleanup before starting and
* atomically registering work. A throwing starter leaves nothing registered;
* after it returns, registration cannot fail. Settlement records the outcome,
* notifies listeners, and releases waiters.
* @param spec - task identity, owner, and synchronous starter.
* @returns the registry-issued `<kind>-N` id.
*/
abstract start(spec: TaskStart): TaskId
/**
* List caller-owned and unowned tasks in registration order without exposing
* another session's labels.
* @param caller - reading agent; a non-agent caller sees only unowned tasks.
* @returns fresh snapshots.
*/
abstract list(caller?: Agent): TaskSnapshot[]
/**
* Return a non-consuming snapshot without changing its read cursor or notice
* state. Throws for an unknown or foreign task.
* @param id - task to look up.
* @param caller - reading agent checked against the owner.
* @returns a fresh snapshot.
*/
abstract get(id: TaskId, caller?: Agent): TaskSnapshot
/**
* Read the next stream delta, or the idempotent final output after settlement.
* A terminal read marks the task reported. Throws for an unknown or foreign
* task.
* @param id - task to read.
* @param caller - reading agent checked against the owner.
* @returns output text and the post-read snapshot.
*/
abstract read(id: TaskId, caller?: Agent): TaskRead
/**
* Request cancellation, then mark the task stopping and reported. A producer
* throw propagates without changing task state. Throws for an unknown or
* foreign task.
* @param id - task to cancel.
* @param caller - killing agent checked against the owner.
* @param reason - logged reason forwarded to the producer.
* @returns `requested` for live work, otherwise `already-finished`.
*/
abstract kill(id: TaskId, caller?: Agent, reason?: string): 'requested' | 'already-finished'
/**
* Wait for settlement or timeout without cancelling the task. Caller abort
* rejects only while the task is live; after settlement the terminal
* snapshot wins so a notice suppressed for this waiter is still delivered.
* Throws for invalid, unknown, or foreign input.
* @param id - task to wait for.
* @param timeoutMs - positive finite wait bound in milliseconds.
* @param caller - waiting agent checked against the owner.
* @param signal - optional cancellation of the wait itself.
* @returns snapshot at settlement or timeout.
*/
abstract wait(id: TaskId, timeoutMs: number, caller?: Agent, signal?: AbortSignal): Promise<TaskSnapshot>
/**
fix(tasks-local): layer control surfaces and listeners by registering scope One host registry serves every composition in the process, so its two service-wide collections answered per-owner questions process-wide. `start()` asked only whether SOME surface was attached, so an agent whose own composition loads no `tool-tasks` could start work it has no tool to collect or stop as soon as any other preset attached one — and the answer changed depending on which sessions happened to be open. `settle()` walked every registered listener, so a task settling without a waiter injected one completion notice per mounted preset into the same owner. Both collections now sit in `ScopedLayers`, the layered-registry primitive `tools` and `skills` already use: a registration files into its registering context's scope, and a read unions the global layer with the owner's scope chain. A surface or listener registered from an unscoped context lands in the global layer and serves every owner, which is exactly the host-plane composition's own controls, so the TUI path is unchanged without a special case. This supersedes the consumer-side filter in the previous commit. That filter produced the right notices but sat in the wrong layer: it left the `start()` gate process-wide, it could not be enforced against a producer that resolves the registry directly, and it made a Consumer carry scope knowledge that the other layered registries keep in the registry. `tool-tasks` is scope-agnostic again and the `dsh-scope` edge moves to `tasks-local`. `start()`'s refusal is now owner-relative, so its model-visible text names the agent rather than the process. The shipped `minimal` preset keeps `enableRunInBackground: false`, no longer as the safety boundary — the registry owns that now — but so an agent that could never collect a task is not offered the parameter at all. Refs #2141
2026-08-10 16:12:41 +08:00
* Register an effect-scoped completion listener. It receives the settlements
* of the owners its registering context's scope covers; each listener is
* contained; returned promises are observed but not awaited. No listener runs
* after service disposal.
* @param listener - receives each terminal snapshot and its exact owner.
* @returns disposer that unregisters the listener.
*/
abstract onTaskDone(listener: TaskDoneListener): () => void
/**
/**
* Register an effect-scoped observer of visible-set changes. It fires after
* every commit that changes what {@link list} returns for that owner —
fix(tasks): correct the change-feed contract and its documentation Review found the `onTasksChanged` teardown reasoning inverted. The comment claimed every registration is an effect on the registry's own fiber, so listeners would be gone before service disposal empties the store — but the traceable proxy rebinds `this.ctx` to the CALLER, which this package's own HMR-safety test already proves. The only shipped consumer registers from the api-proxy mux stream, so it was still listening and simply kept the rows it last received. Service disposal now announces the emptied set, and teardown announces its stopping transition immediately instead of leaving an observer on `running` for however long a slow producer takes to release. Two documentation claims were false in the opposite direction: the Agent Note and the ui-task README both said an unowned task is invisible in the header, while `list(caller)` returns unowned tasks to every caller, the carrier fans their changes out to every subscribed session, and this PR's own test asserts exactly that. The note even contradicted itself two sections earlier. Both sides now state the real asymmetries — another session's tasks, and the process-local registry emptying on restart. The "no Web path calls the consuming `ctx.tasks.read()`" invariant claimed a test that did not exist; the carrier suite's producer had no `readOutput` at all, so a stray read would have failed nothing. Its producer now counts cursor consumption and the lifecycle and baseline paths both assert zero. Also: a session created after the mux opened now receives the task baseline it missed, the popover samples its clock when it opens rather than at mount, and a failed task's unbounded producer detail elides instead of widening the row.
2026-08-10 13:37:18 +08:00
* registration, every stopping transition (including the one teardown
* performs before it awaits a slow producer), settlement, owner-disposal
* removal, and the emptying that service disposal commits — so an observer
* re-reads rather than accumulating deltas.
*
* Delivery is owner-relative on the same terms as {@link onTaskDone}: an
* observer registered from an unscoped context — a host composition's own
* carrier — sees every owner, while one registered under an agent
* composition's scope sees exactly the agents composed under it.
*
* This is not a superset of {@link onTaskDone}: that one delivers the terminal
* record under first-wins semantics a task controller couples to notice
* delivery, while this one carries no delivery meaning and marks nothing
* reported. Listeners are contained and never awaited.
* @param listener - receives the owner whose visible set changed, or
* `undefined` when an unowned task changed and every caller's set did.
* @returns disposer that unregisters the listener.
*/
abstract onTasksChanged(listener: TasksChangedListener): () => void
/**
* Attach an effect-scoped controller that can read and stop tasks. It serves the
fix(tasks-local): layer control surfaces and listeners by registering scope One host registry serves every composition in the process, so its two service-wide collections answered per-owner questions process-wide. `start()` asked only whether SOME surface was attached, so an agent whose own composition loads no `tool-tasks` could start work it has no tool to collect or stop as soon as any other preset attached one — and the answer changed depending on which sessions happened to be open. `settle()` walked every registered listener, so a task settling without a waiter injected one completion notice per mounted preset into the same owner. Both collections now sit in `ScopedLayers`, the layered-registry primitive `tools` and `skills` already use: a registration files into its registering context's scope, and a read unions the global layer with the owner's scope chain. A surface or listener registered from an unscoped context lands in the global layer and serves every owner, which is exactly the host-plane composition's own controls, so the TUI path is unchanged without a special case. This supersedes the consumer-side filter in the previous commit. That filter produced the right notices but sat in the wrong layer: it left the `start()` gate process-wide, it could not be enforced against a producer that resolves the registry directly, and it made a Consumer carry scope knowledge that the other layered registries keep in the registry. `tool-tasks` is scope-agnostic again and the `dsh-scope` edge moves to `tasks-local`. `start()`'s refusal is now owner-relative, so its model-visible text names the agent rather than the process. The shipped `minimal` preset keeps `enableRunInBackground: false`, no longer as the safety boundary — the registry owns that now — but so an agent that could never collect a task is not offered the parameter at all. Refs #2141
2026-08-10 16:12:41 +08:00
* owners its registering context's scope covers, and {@link start} refuses an
* owner no attached controller serves.
* @param name - diagnostic label; duplicate names remain independent.
* @returns disposer that detaches this controller.
*/
abstract attachController(name: string): () => void
```
Types: [Agent](core.md)
Source: [`packages/tasks/tasks/src/index.ts:62`](../../packages/tasks/tasks/src/index.ts)
<!-- END GENERATED cordis-surface -->