deepseek-harness/docs/architecture.md

DeepSeek Harness Architecture

English | 中文

DeepSeek Harness SDK uses Cordis: everything is a plugin, including the loop.

Overview

Harnesses are Cordis contexts; packages contribute services, typed events, and disposable registrations.

packages/core/ groups the default agent flow; capabilities remain plugins.

Default Services

ctx key	Package	Role
—	dsh-scope	scoped-context registrations and shared layer storage (library)
ctx.sessions	dsh-session	in-memory event-sourced sessions
ctx.systemPrompt	dsh-system-prompt	ordered prompt sections, tool schemas, and variables
ctx.tools	dsh-tools	tool registry and execution pipeline
ctx.agents	dsh-agent	live agents, delegated creation, agent/* events, process-local initiator scope
ctx.agentLoop	dsh-agent-loop	concrete Agent driver

Capability Services

ctx key	Package family	Role
ctx.llm	llm/	adapter registry, streaming model calls
ctx.tokenMeter	llm/token-meter	replay-aware request and surface pressure
ctx.bash	bash/	foreground/background command execution
ctx.subprocess	subprocess/	managed child-process trees for bash, LSP, and ACP subagent backends
ctx.pty	pty/	owner-scoped persistent terminal sessions
ctx.sandbox	sandbox/	same-world process confinement through argv wrapping and per-call policy
ctx.sandboxPolicy	sandbox/	shared sandbox policy home
ctx.codeRuntime	code-runtime/	model-written program execution
ctx.fs	fs/	filesystem provider primitives and policy events
ctx.lsp	lsp/	semantic navigation registry
ctx.skills	skill/	skill provider registry, progressive disclosure
ctx.web	web/	search/fetch provider registries
ctx.compact, ctx.toolResultPrune	compact//compact-tool-result-prune	summary compaction, optional model-free result pruning
ctx.subagents	subagent/	named delegation providers
ctx.planMode	plan/	logged plan collaboration state
ctx.tasks	tasks/	background task registry, generic task_* controls
ctx.workflows	workflow/	script-driven multi-agent orchestration
ctx.goals	goal/	persisted same-session goals
ctx.sessionPersistence	session-persistence/	durable session-log storage
ctx.sessionQuery	session-query/	live-preferred exact/filter/trace queries over SQLite FTS, workspace-authorized model tools
ctx.sessionTitle	session-title/	log-backed fallbacks, one optional asynchronous provider
ctx.directoryPicker	host/directory-picker	GUI-host directory picking (native/browse interactions)
ctx.invariants	support/invariants	package-name-selected registry of package-owned runtime checks

Event

Events are the service extension API (catalog, producer/consumer map).

Event Domains

• Session events are durable log facts emitted through session/event.
• Agent events carry live Agent for status, prompt admission, request shaping, validation, and continuation.
• Capability events let owning seams attach policy and adapters without a loop import.

Interception Semantics

Waterfalls are around-middleware: listeners delegate with next(); returning without it vetoes or takes over (semantics).

Default Loop Lifecycle

A session is append-only. An ordinary turn claims one queued follow-up; injection claims none. A successor awaits its predecessor's checkpoint but may share its running interval (decision). A turn ends when model or plugins stop it; a step is one model request plus tools. Quotes in the sequence below mark durable events.

Creation without an id mints <config-id>-session-<uuid>; sessionId resumes or creates, while resumeSessionId requires history. Resume restores lineage and delegation depth before publication. Setup failures emit agent-loop/config-start-failed; teardown is silent.

Turn Flow

choose declarative identity and fresh/resume path
  -> prepare private session + agent.ctx -> await unpublished setup
  -> enter session + agent -> session/created -> agent/created
  -> enable driving -> agent/session-start(source) -> start driver
forever:
  wait for a queued message
  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
      append a context-only caller batch immediately
      keep steering and context staged beside it pending for a later admitted turn
    allowed prompt:
      'turn/start'
      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 tool schemas
      snapshot the derived messages (the reconstruction boundary)
      'step/start'
      agent/request (config only) -> prepare reasoning/default under turn signal -> log request/header -> llm/stream (frozen, registration-bound)
      'assistant/chunk'
      'assistant/message'
      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'
      drain accepted tool context and steering
      '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)

idle inject:
  append 'user/message'
  do not open a turn or run the model

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).

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. 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, retry foundation, provider policy).

Failure Boundaries

Final-adapter selection, dispatch, and iteration failures become terminal finish { kind: 'error' | 'aborted', failure } chunks before the loop handles them. agent/request-error receives request coordinates, normalized LlmFailure, the prepared registration's retry policy when available, and the signal; middleware and consumer errors remain thrown outside request recovery. Failed chunks commit neither messages nor tool calls.

Other failures use agent/error. Cancellation and disposal beat recovery. Before request-header commit, the turn signal cancels asynchronous model-capability preparation; undispatched tools get synthetic tool/call/ABORTED_BEFORE_DISPATCH pairs. Effective cancel(cause) emits its cause before queue clearing and abort; observers cannot veto; idle calls emit nothing. Durability records user or parent cancellation as aborted, teardown as disposed; teardown awaits quiescence. The cause affects reporting, not late result-context handling (decision).

Turn and step events are turn-enclosed; idle injected user/message events may sit between turns. Reload closes an interrupted tail with a synthetic turn end. After close, only agent/error reports failures. Each turn has one TurnEndReason.

Agent Handles

ctx.agents returns AgentHandle { agent, dispose() }. Plugins drive agents with followup(), steer(), and inject(); cancel() stops work, while the awaited disposer owns teardown.

Agent Scope

Each agent owns scoped agent.ctx; shared storage overlays its tool, prompt, and command entries on globals while preserving domain views (decision). Scoped listeners filter dispatch; contributions unwind with awaited cleanup. CreateAgentOptions.setup(agentCtx) composes before publication. Typed resolvers derive carrier checks from merged Events and scopeTarget (semantic gates). Details: agent scope, subagent composition. AgentLoop runs under ctx.agents.withInitiator(); private orchestration derives agent.session, but turn, step, signal, cwd, and authority stay explicit (decision).

State

Session Log

The session log is authoritative. deriveMessages() projects model history; raw assistant/chunk events preserve replay and UI fidelity. Fork, resume, transcript rendering, telemetry, and persistence derive from this stream.

Model-visible ⟺ logged: messages at step/start plus the folded request/header reconstruct every request; package-owned dsh-agent-loop/invariant can assert this through ctx.invariants (reconstructability).

Durability is a plugin concern. Backends eagerly drain synchronous session/event notifications. session/flush barriers precede each request and top-level tool dispatch, then follow turn/end before another queued turn or idle observation. SessionPersistence stores SessionEvent directly and metadata in SessionHeader; JSONL defaults to checksummed Zstandard, while SQLite shares the contract (decision).

Log-only events may sit between turns. Owners append through Session, flushing only for durability. session/title relies on eager persistence and lifecycle drains. Latest title wins with provenance; fallback and provider work never delays responses. Such records are fork boundaries, so forks inherit titles (decision).

Model Content

Messages use typed blocks from merge-extensible ContentBlockMap; the pattern also types MessageSource, FinishReason, TurnTrigger, and TurnEndReason. New blocks coordinate adapters, UI, compaction, token metering, and persistence; replay measurements live in token-meter.md.

Streaming uses raw chunks and BlockAssembler. Each LlmAdapter.stream() is one provider attempt; adapters report normalized failure facts, and a handling agent/request-error plugin returns a retry action. The loop logs chunks, successful provenance, and replay state. Remote adapters use per-read idle watchdogs. Replay crosses routes only through a shared adapter instance (contract).

Extension And Composition

Capability Pattern

A swappable capability usually has interface / implementation / consumer layers: service/events, backend, and model-facing tools/prompts. Bash is the reference; the capability graph maps each family.

Exceptions combine LLM interface/consumer, filesystem policy, web registries, and named skill/subagent providers. Subagents spawn fresh, fork a completed-turn prefix, or use ACP children (subagent.md).

dsh-workspace-context injects baseline at the first agent/step and appends ctx.fs-discovered changes through tools/post-execute; its decision records isolation. dsh-paths owns shared paths.

Bundles And Apps

dsh-agent-spine-demo bundles a spine and optional goals. App packages own TUI, CLI, ACP automation, and JSON-RPC front doors (README, acp/, ui/). dsh-jsonrpc-agent boots external cordis.yml; the Python SDK defaults when config is absent (Python SDK). Thin deployments use swappable backends and optional tools (examples/, runnable wirings, graph atlas).

Where New Behavior Goes

New behavior attaches to a documented extension point; a loop change updates this map.

Goal	Mechanism
Add a model provider	register its adapter on ctx.llm
Add a model-facing capability	register on ctx.tools; schemas join prompt assembly
Add shell execution	implement and register a ctx.bash backend; the local backend spawns through ctx.subprocess
Add persistent terminal execution	register a ctx.pty backend plus dsh-tool-pty
Add a human command	register on ctx.commands; adapters discover and dispatch without a model turn
Add background work	register on ctx.tasks; generic task_* tools collect or stop it
Add filesystem access or policy	implement a ctx.fs provider or listen to fs/* policy events
Confine spawned processes	use a ctx.sandbox backend; consumers wrap argv before spawning
Intercept a request, tool, or turn	use its agent/* or tools/* event; agent/turn-stopping is the stop boundary
Add model-facing context	call agent.inject() to append a sourced user/message without a turn
Add UI or editor integration	drive ctx.agents, render from session/event; terminal-only overlays use ctx.tui
Add durable session state	extend SessionEventMap; render and replay from the log
Add asynchronous session-title generation	register the sole ctx.sessionTitle provider
Manage a same-session objective	use ctx.goals; continue through Agent and agent/*
Fork a live session	call ctx.sessions.fork(source, boundary?, childSessionId?)
Scope a registration to one agent	use its agent.ctx (see Agent Scope)

The extension cookbook has plugin skeletons and the feature-to-seam map; guides cover packages, tools, LLM adapters, and vendored packages.