Coder/deepseek-harness
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Merge latest agent execution context base

Browse Source
This commit is contained in:
Yichen Jiang
2026-07-18 21:42:38 +08:00
parent cddb79e844 0e4785cdb7
commit 5190e10d27
105 changed files with 3313 additions and 551 deletions
Download Patch File Download Diff File Expand all files Collapse all files

17
docs/architecture.md
View File

@@ -84,11 +84,12 @@ forever:
'assistant/chunk'
agent/step-result
'assistant/message' (transformed content or empty success anchor after step-result rejection)
each tool call:
'tool/call'
tools/pre-execute -> monotonic guards -> tools/execute -> tools/post-execute -> tools/result
'tool/result'
append post-tool context and steering
schedule tool calls by ctx.tools.executionMode:
exclusive -> one-call barrier
parallel -> rolling pool, <= maxParallelToolCalls in flight; reclassify before start
each start -> 'tool/call' -> ordered tools/pre-execute -> concurrent tools/execute
each model-order result -> ordered tools/post-execute -> 'tool/result'
append accepted tool-batch context after all recorded results, then steering
'step/end'
agent/turn-continuation
agent/turn-stop (terminal policy)
@@ -99,11 +100,11 @@ forever:
Each step assembles ordered prompt sections, tool schemas, and `{{name}}` variables; unknown or valueless references fail the turn. `dsh-system-prompt` owns the harness identity and default persona, which an agent scope may shadow. The loop supplies `model` and `cwd` ([prompt-ownership RFC](rfc/implemented/architecture/2026-07-05-prompt-variables-and-tool-guidance-ownership.md)).
Post-tool context lands after all tool results so tool-call/result adjacency stays stable. Steering drains between steps; ordinary leftover steering after a turn is re-queued as input. A terminal `agent/turn-stop` is the explicit exception: it runs after ordinary continuation and steering folding, then remains authoritative through turn close and flush so steering from those later listeners is discarded rather than becoming another step or turn; ordinary queued prompts are preserved.
Context accepted during tool execution—including async `agent.inject()` notices and post-tool `additionalContexts`—waits for settlement, then follows every recorded result. Successful batches preserve call/result adjacency; interrupted batches drain that context before turn closure. Steering drains between steps; ordinary leftover steering becomes queued input. Terminal `agent/turn-stop` runs after continuation and steering folding, stays authoritative through turn close and flush, and discards later steering while preserving queued prompts.
### Failure Boundaries
The turn is the containment boundary. A throwing listener, adapter error finish, or failed step ends it with an error reason and reports `agent/error` without killing the driver. One explicit `AbortSignal` spans prompt submission, prompt assembly, every step, model and tool work, continuation, turn close, and flush. `cancel()` clears queued and steering work and carries a typed `user` or `parent` runtime cause, while the durable turn records only `aborted`; disposal is a distinct higher-priority terminal state that awaits quiescence before unregistering the agent and draining service disposers. Cooperative work must settle before the loop reports quiescence. See the [explicit turn cancellation decision](rfc/implemented/architecture/2026-07-16-explicit-turn-cancellation.md).
The turn contains failures: a throwing listener, adapter error finish, or failed step records an error and reports `agent/error` without killing the driver. One explicit `AbortSignal` spans prompt submission, assembly, steps, model/tool work, continuation, turn close, and flush. `cancel()` clears queued and steering work and carries a typed `user` or `parent` cause; the durable turn records only `aborted`. Disposal is a distinct, higher-priority terminal state that waits for quiescence before unregistering the agent and draining service disposers. Cooperative work must settle before quiescence. See the [explicit turn cancellation decision](rfc/implemented/architecture/2026-07-16-explicit-turn-cancellation.md).
Every session event is turn-enclosed. Reloading preserves an interrupted tail and closes it with a synthetic `interrupted` turn end. Failures after durable turn close report only through `agent/error` because no safe in-turn position remains. Each turn has one `TurnEndReason`; [TurnEndReasonMap](core-data-structures/session.md#why-a-turn-ended-turnendreasonmap) owns the variants.
@@ -117,7 +118,7 @@ Every live agent owns a scoped `agent.ctx`. Its registrations shadow globals, re
### Agent Execution Context
`AgentLoop` wraps each concrete driver in process-local `ctx.agentExecution`; its ALS frame contains only `{ agent }`. Child creation and setup stay outside the boundary, and turn, step, signal, cwd, and authority remain explicit. See the [package contract](../packages/core/agent-execution/README.md) and [decision](rfc/implemented/architecture/2026-07-15-agent-execution-context.md).
`AgentLoop` wraps each driver in process-local `ctx.agentExecution`; its ALS frame is exactly `{ agent }`. Child creation and setup stay outside, while turn, step, signal, cwd, and authority remain explicit. See the [package contract](../packages/core/agent-execution/README.md) and [decision](rfc/implemented/architecture/2026-07-15-agent-execution-context.md).
## State