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

docs(agent): document next-step acceptance window

Browse Source
This commit is contained in:
_Kerman
2026-07-27 18:07:51 +08:00
parent f63d937496
commit ff7792b17c
35 changed files with 267 additions and 206 deletions
Download Patch File Download Diff File Expand all files Collapse all files

11
docs/architecture.md
View File

@@ -76,10 +76,12 @@ choose declarative identity and fresh/resume path
-> enable driving -> agent/session-start(source) -> start driver
forever:
wait for a queued message
claimed message -> agent/prompt-submit
blocked prompt -> park without opening a turn
claim message -> emit agent/status(running) if starting an interval
open the next-step acceptance window
-> agent/prompt-submit
blocked or failed prompt -> close the window without opening a turn
caller-staged context and steering remain pending for a later admitted turn
allowed prompt:
emit agent/status(running)
'turn/start'
append prompt + additional contexts as separate 'user/message' events
STEP loop:
@@ -100,6 +102,7 @@ forever:
'step/end'
continue for tools or steering unless a result concluded the turn
otherwise agent/turn-stopping -> drain -> 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)
@@ -110,7 +113,7 @@ idle inject:
Each step assembles ordered prompt sections, 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)).
Tool-time context—active-turn `inject()` and post-tool `additionalContexts`—settles after results; steering drains there and requests another step. Idle `inject()` appends immediately without changing turn numbers; persistence drains eagerly.
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. Idle `inject()` appends immediately without changing turn numbers; persistence drains eagerly.
Pruning precedes summaries; overflow retries require durable progress. Recovery uses `agent/request-error` after the failed step and before turn close. Its policy calls `agent.retry()` to schedule one retry turn; cancellation wins ([compaction](../.agents/notes/implemented/architecture/2026-07-10-after-call-compaction-pressure-and-overflow-recovery.md), [retry](../.agents/notes/implemented/architecture/2026-06-21-bounded-llm-request-recovery.md)).