deepseek-harness/docs/agent-lifecycle.md

<!-- Generated by scripts/gen-doc-graphs.ts - do not edit by hand.
     Run `pnpm run gen-doc-graphs` to regenerate. -->

# Agent Turn And Step Lifecycle

This sequence is the visual companion to [architecture.md](architecture.md#default-loop-lifecycle). It keeps durable replay facts on `session/event` and live control/status on `agent/*`.

```mermaid
sequenceDiagram
  participant User
  participant Agent
  participant Driver
  participant Hooks as hook listeners
  participant Prompt as ctx.systemPrompt
  participant LLM as ctx.llm
  participant Tools as ctx.tools
  participant Session
  participant SDK as UI or SDK listener
  User->>Agent: followup(content)
  Agent-->>SDK: <code>agent/inbox/spliced</code>
  Agent-->>SDK: <code>agent/inbox/inserted</code> { message }
  Agent->>Driver: queued work wakes driver
  Driver-->>SDK: <code>agent/status</code> running
  Note over Agent,Driver: claim pending next-step input plus one queued prompt
  Driver-->>SDK: <code>agent/inbox/spliced</code> pure deletion
  Driver-->>SDK: <code>agent/inbox/claimed</code> { message, turn } per message
  Driver->>Hooks: <code>agent/pre-step</code> waterfall
  Hooks-->>Driver: authoritative reject or enter(messages)
  alt proposed step rejected or pre-step failed
    Driver-->>Driver: claimed batch stays removed, no turn opens
  else enter proposed step
  Driver->>Session: <code>turn/start</code>
  Driver->>Session: <code>step/start</code>
  Driver->>Session: <code>user/message</code> per entered message
  Driver->>Prompt: <code>system-prompt/assemble</code> waterfall
  Driver->>LLM: <code>agent/request</code> waterfall, then <code>llm/stream</code> waterfall
  LLM-->>Driver: StreamChunk*
  Driver->>Session: <code>assistant/chunk</code>*
  Session-->>SDK: <code>session/event</code> <code>assistant/chunk</code>*
  alt final adapter or terminal in-band request failure
    Driver->>Session: <code>step/end</code>
    Driver->>Hooks: <code>agent/request-error</code> waterfall
    Hooks-->>Driver: return retry action or preserve the original error
  else model request succeeded
  Driver->>Session: <code>assistant/message</code>
  Driver->>Tools: classify pending call by executionMode
  loop barriers and bounded rolling pool, reclassify before start
    opt call starts
      Driver->>Session: <code>tool/call</code>
      Driver->>Tools: ordered pre, concurrent execute
      Tools-->>Session: tool-owned events when applicable
    end
    opt next model-order result ready
      Driver->>Tools: ordered post
      Driver->>Session: <code>tool/result</code>
    end
  end
  Driver->>Session: <code>step/end</code>
  opt natural stop and next-step inbox empty
    Driver->>Hooks: <code>agent/turn-stopping</code> serial terminal checkpoint
  end
  opt next-step input is pending
    Driver-->>Driver: claim pending next-step input
    Driver-->>SDK: <code>agent/inbox/claimed</code> { message, turn } per message
    Driver->>Hooks: <code>agent/pre-step</code> waterfall
    Hooks-->>Driver: authoritative reject or enter(messages)
  end
  end
  Driver->>Session: <code>turn/end</code>
  end
  Driver-->>SDK: <code>agent/status</code> idle
```

The `assistant/message` event records every successful provider call, including content-less and `max-tokens` finishes. Empty content stays out of derived history, while the durable event keeps usage and `sourceEventSeqs` listing the exact `assistant/chunk` events, including an explicit empty list.

`dsh-compact-basic` uses `agent/pre-step` for pressure before request derivation and `agent/request-error` only for canonical context overflow. Once either trigger qualifies, optional tool-result pruning runs before summary selection. Recovery works between the closed failed step and failed turn close, and opens a fresh retry turn only when pruning or summarization advances the surface replacement generation; otherwise the original request error remains authoritative.

The returned `agent/pre-step` decision is authoritative; listeners wrapping `next()` preserve downstream messages unless replacement is intentional. Steering and injected context pass through the same waterfall after a later claim operation takes their next-step batch.

SDK users that need replayable transcript data should consume `session/event`; `agent/*` is the live coordination surface for queue/status, prompt interception, request construction, steering, continuation, and errors.

Maintenance mode: curated Mermaid sequence; exact event signatures live in the generated Cordis catalog.