Files
deepseek-harness/packages/workflow
Yichen Jiang 483199d47a Merge branch 'worktree-llm-dynamic-config' into worktree-llm-web-config
# Conflicts:
#	apps/cli/cordis.yml
#	apps/web/tests/snapshots/code-mode-round/session.jsonl
#	apps/web/tests/snapshots/cordis-tool-round/session.jsonl
#	apps/web/tests/snapshots/fresh-round-trip/session.jsonl
#	apps/web/tests/snapshots/lifecycle-chrome/session.jsonl
#	apps/web/tests/snapshots/live-interactions/session.jsonl
#	apps/web/tests/snapshots/navigation-panes/seed.jsonl
#	apps/web/tests/snapshots/question-composer/session.jsonl
#	apps/web/tests/snapshots/seeded-history/seed.jsonl
#	apps/web/tests/snapshots/steering/session.jsonl
#	docs/cordis-catalog/events.md
#	docs/cordis-catalog/services.md
#	docs/core-data-structures/core.i18n.yaml
#	docs/core-data-structures/settings.i18n.yaml
#	docs/event-producer-consumer.md
#	docs/module-graph.md
#	examples/acp-agent/tests/snapshots/workspace-context/session.jsonl
#	packages/client/connection/README.i18n.yaml
#	packages/client/connection/src/index.ts
#	packages/client/connection/tests/node-half.spec.ts
#	packages/client/runtime/README.i18n.yaml
#	packages/client/runtime/README.md
#	packages/client/runtime/README.zh.md
#	packages/client/runtime/src/client/index.ts
#	packages/client/runtime/tests/fake-api.ts
#	packages/client/ui-models/README.i18n.yaml
#	packages/examples/tui-demo/README.i18n.yaml
#	packages/host/apiproxy/README.i18n.yaml
#	packages/host/apiproxy/package.json
#	packages/host/apiproxy/src/api-proxy.ts
#	packages/host/apiproxy/src/api/rpc.schema.ts
#	packages/host/apiproxy/src/api/rpc.ts
#	packages/llm/llm-deepseek/README.i18n.yaml
#	packages/llm/llm-deepseek/README.zh.md
#	packages/llm/llm-pi-ai/README.i18n.yaml
#	packages/llm/llm/README.i18n.yaml
#	packages/llm/llm/README.zh.md
#	packages/sdk/sdk-client/README.i18n.yaml
#	packages/settings/settings/README.i18n.yaml
#	packages/settings/settings/README.md
#	packages/settings/settings/README.zh.md
#	packages/subagent/subagent-dsh-sdk/README.i18n.yaml
#	packages/subagent/subagent-dsh-sdk/README.zh.md
#	packages/support/llm-replay/README.i18n.yaml
#	packages/ui/jsonrpc/README.i18n.yaml
#	packages/ui/jsonrpc/README.zh.md
#	packages/ui/tui/tests/snapshots/model-selector.expected.txt
#	packages/ui/tui/tests/snapshots/model-switching.expected.txt
#	packages/ui/tui/tests/snapshots/resume-sessions.expected.txt
#	packages/ui/tui/tests/snapshots/status-diagnostics-narrow.expected.txt
#	packages/ui/tui/tests/snapshots/status-diagnostics.expected.txt
#	packages/ui/tui/tests/tui.snapshot.ts
#	pnpm-lock.yaml
#	python/sdk/README.i18n.yaml
#	scripts/snapshots/translation-prompt-v4/request-response.expected.json
2026-07-30 15:18:26 +08:00
..
2026-07-26 05:06:39 +08:00

workflow/ — dynamic-workflow capability family

English | 中文

The workflow seam: a model-written JavaScript orchestration script that fans out subagents at scale (phases, structured per-agent results, concurrency caps), modeled on Claude Code's dynamic workflows. A capability seam (see capability seams) in the bash shape: ONE engine implementation per context registers as ctx.workflows; the model-facing tool consumes it.

Package Role ctx key
workflow/ Abstract workflow seam: service base class + run vocabulary + workflow/* events ctx.workflows
workflow-workerthread/ node:worker_threads engine: one worker per run; the script's vm context lives inside the worker, agent() bridges to ctx.subagents over the message port (provides ctx.workflows)
tool-workflow/ Model-facing workflow tool over ctx.workflows (registers on ctx.tools)
tool-ralph/ Fixed fresh-agent Ralph policy over ctx.workflows and a fresh structured-output subagent provider (registers on ctx.tools)

The interface lives at workflow/workflow/. The engine's agent() hook rides the subagent seam (any registered provider; the shipped examples use spawn), and agent({ schema }) rides the structured-output support the in-process backends implement. The worker thread isolates the SCRIPT — the host never blocks on it, and a cancelled run's post-grace termination is real — but it is NOT a security boundary; an isolated-vm/separate-process engine (actual sandboxing) swaps in behind the same interface if that ever matters.

The general script engine's decisions and deferred work live in the dynamic-workflows Agent Note. The separate Ralph consumer fixes the script and fresh-provider policy rather than adding another engine or an agent-loop mode.