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

Merge origin/master into skill system branch

Browse Source 
Resolve documentation split, tool presentation, and generated catalog changes from master while preserving the skill system integration.
This commit is contained in:
Yichen Jiang
2026-07-05 16:50:29 +08:00
parent 5870bc2e1a 57fd914f95
commit 83dfc3887e
454 changed files with 29524 additions and 4598 deletions
Download Patch File Download Diff File Expand all files Collapse all files

22
packages/core/agent-loop/README.md
View File

@@ -46,10 +46,14 @@ Agents listed in config are auto-created at startup. `cwd` applies only to fresh
One invocation of `runLoop()` drives one agent for its whole lifetime:
```
create agent → emit agent/session-start(source) ⟵ once, before turn 1
forever:
wait for queued messages (idle)
TURN (error-contained):
drain queued → 'turn/start' → session('user/message')
'turn/start'
each queued: waterfall agent/prompt-submit → allow (→ session('user/message'),
inject additionalContext) | block (→ session('prompt/blocked'), drop)
if every prompt blocked: 'turn/end'(rejected), no step ⟵ zero-step turn
STEP loop:
drain steering
assembly = systemPrompt.assemble()
@@ -59,10 +63,14 @@ forever:
stream llm.stream(request) → session('assistant/chunk')
message = waterfall agent/step-result
session('assistant/message')
each tool-call: session('tool/call') → tools.execute() → session('tool/result')
each tool-call: session('tool/call')
→ tools.execute() [waterfall tools/pre-execute → dispatch → tools/post-execute]
→ session('tool/result')
append buffered post-execute additionalContext as session('context/message')(s)
drain steering → session('steering/message')
cont = waterfall agent/turn-continuation
if !cont: break
cont = waterfall agent/turn-continuation → ContinuationDecision
({action:'continue', reason?} records reason as next-step steering)
if action==stop (and no pending steering): break
session('turn/end')
await session/flush
re-enqueue leftover steering as queued
@@ -76,9 +84,9 @@ Cancellation: `agent.cancel()` is the single public stop primitive — it clears
### What is NOT here
Everything that goes beyond "call the model, run the tools, repeat" belongs to plugins listening on the event taxonomy:
- Hooks: `agent/pre-step`, `agent/request`, `agent/step-result`, `tools/execute`, `agent/turn-continuation`
- Hooks: `agent/session-start`, `agent/prompt-submit`, `agent/pre-step`, `agent/request`, `agent/step-result`, `tools/pre-execute`, `tools/post-execute`, `agent/turn-continuation`
- Compaction: `agent/pre-step`
- Sandbox, permission, plan mode: `tools/execute`
- Sandbox, permission, plan mode: `tools/pre-execute` (deny/ask gate), `tools/post-execute`
- Sub-agents: implemented outside the loop as `ctx.subagents` providers; in-process providers use `ctx.agents.create()` and owned `AgentHandle` teardown, while child streaming/progress and background/poll collection remain deferred.
- Persistence: `session/event` + `session/flush`
- UI: `agent/stream-chunk` + `agent/*` events
- UI: `session/event` (assistant token stream, boundaries, tool activity) + `agent/*` control events (`agent/status`, `agent/created`/`agent/disposed`)