DeepSeek/deepseek-harness
1
0
Fork 0
mirror of https://github.com/deepseek-ai/deepseek-harness synced 2026-08-15 21:04:50 +00:00
Code Issues Packages Projects Releases Wiki Activity

fix(subagent): confirm steering request admission

Browse Source
This commit is contained in:
Dudu-0223
2026-07-24 14:32:19 +08:00
committed by imccyu
parent 189502e4ac
commit e1f7eeeb95
58 changed files with 565 additions and 480 deletions
Download Patch File Download Diff File Expand all files Collapse all files

14
docs/architecture.md
View File

@@ -93,12 +93,12 @@ forever:
append prompt + additional contexts as separate 'user/message' events
STEP loop:
agent/step
drain injected context and steering (steering bypasses prompt-submit)
assemble system prompt and tools
materialize changed runtime context as sourced 'user/message'
drain injected context and provisional steering (steering bypasses prompt-submit)
snapshot the derived messages (the reconstruction boundary)
'step/start'
open strict-steering acceptance
admit the drained steering receipts
agent/request (config only) -> prepare adapter defaults/provenance + context capacity under turn signal -> log request/header (+ request/context on route change) -> llm/stream (frozen, registration-bound)
'assistant/chunk'
'assistant/message'
@@ -107,10 +107,10 @@ forever:
parallel -> rolling pool, <= maxParallelToolCalls; reclassify-at-start; scheduler failure -> stop starts, drain dispatches
start -> 'tool/call' -> ordered tools/pre-execute -> concurrent tools/execute
model-order result -> ordered tools/post-execute -> 'tool/result'
close strict-steering acceptance, then drain accepted tool context and steering
drain accepted tool context after all results; keep steering provisional
'step/end'
continue for tools or steering unless a result concluded the turn
otherwise agent/turn-stopping -> drain -> continue only for steering
continue for tools or steering unless a result concluded the turn and rejects pending steering
otherwise agent/turn-stopping -> drain context -> continue only for steering
close the next-step acceptance window
'turn/end' -> agent/settled
start the next waking queued message, or emit agent/status(idle)
@@ -122,7 +122,7 @@ idle inject:
Each step assembles ordered stable system sections, cache-safe dynamic contexts, tool schemas, and variables; unknown references fail the turn. `dsh-system-prompt` owns identity and persona; the loop supplies `provider`, `model`, and `cwd` ([prompt ownership](../.agents/notes/implemented/architecture/2026-07-05-prompt-variables-and-tool-guidance-ownership.md)).
Admission-time and active-turn `inject()` stage for the next step; post-tool `additionalContexts` settles after results. Steering shares that staging boundary and requests another step. The default loop closes its optional `trySteer()` acceptance immediately before the final steering drain; ordinary `steer()` keeps its best-effort routing semantics. Idle `inject()` appends immediately without changing turn numbers; persistence drains eagerly.
Admission-time and active-turn `inject()` stage for the next step; tool-time injection and post-tool `additionalContexts` settle after results. Steering shares the outbox but remains provisional until a request admits it. `steer()` returns a message-owned receipt: after `agent/step` and asynchronous prompt assembly succeed, the loop commits the stable batch, snapshots request history, opens `step/start`, then resolves its receipts as admitted with the turn and step; later arrivals wait. A turn-concluding tool result, broad cancellation, disposal, or a claimed idle-steering turn that never opens a step rejects affected receipts, while `cancel(..., { keepInbox: true })` and non-terminal routing preserve pending delivery. Idle `inject()` appends immediately without changing turn numbers; persistence drains eagerly.
Pruning precedes summaries; overflow retries require durable progress. `agent/request-error` may authorize one retry turn between failed-step and turn close; cancellation wins. Adapter-owned `retryPolicy` makes normal mode bounded; always mode delegates specialized recovery before retrying until success or cancellation ([compaction](../.agents/notes/implemented/architecture/2026-07-10-after-call-compaction-pressure-and-overflow-recovery.md), [retry foundation](../.agents/notes/implemented/architecture/2026-06-21-bounded-llm-request-recovery.md), [provider policy](../.agents/notes/implemented/feature/2026-07-24-provider-retry-policies.md)).
@@ -136,7 +136,7 @@ Turn and step events are turn-enclosed. Idle `user/message` and standalone `comp
### Agent Handles
`ctx.agents` owns agents, returning `AgentHandle { agent, dispose() }`. Plugins use `send()` or `followup()`, `steer()`, optional `trySteer()`, and `inject()` presets; [`reserveTurnAdmission()`](../packages/core/agent/README.md#agent-interface-typests) synchronously reserves idle for durable work without changing queued prompt identity. The default loop's `trySteer()` atomically rejects after the current step's final steering drain begins, while ordinary `steer()` retains best-effort routing. `cancel()` and `whenIdle()` control lifecycle. Awaited disposal owns teardown.
`ctx.agents` owns agents, returning `AgentHandle { agent, dispose() }`. Plugins use `send()` or `followup()`, receipt-bearing `steer()`, and `inject()` presets; [`reserveTurnAdmission()`](../packages/core/agent/README.md#agent-interface-typests) synchronously reserves idle for durable work without changing queued prompt identity. Await a steering receipt when request admission matters; best-effort UI steering may ignore it. `cancel()` and `whenIdle()` control lifecycle. Caller, factory, and consumer co-own teardown through one awaited disposer.
### Agent Scope