Commit Graph

47 Commits

Author SHA1 Message Date
xiaoju 178f6c7519 chore: bump all packages to 0.2.0 2026-05-08 02:55:29 +00:00
xiaomo 3153ab26f6 Merge pull request 'feat(engine): supervisor scene — opt-in LLM thread stop (Phase 3)' (#116) from feat/110-phase3-supervisor into main 2026-05-08 02:45:20 +00:00
xiaoju 014c442ed2 feat(engine): add supervisor scene — opt-in LLM-based thread stop (Phase 3)
Supervisor replaces maxRounds as primary stop mechanism. Every N rounds
(configurable via supervisorInterval, default 3), the engine calls a
cheap LLM to evaluate thread progress and decide continue/stop.

- New engine/supervisor.ts: runSupervisor + parseSupervisorDecisionText
- Supervisor is opt-in: no models.supervisor configured = always continue
- WorkflowConfig gains supervisorInterval (default 3, 0 to disable)
- Engine calls supervisor after each supervisorInterval rounds
- 256 tests pass, 14 new tests for supervisor logic

Refs #110
2026-05-08 02:38:54 +00:00
xingyue 520b17b351 chore: remove all deprecated code
- Remove createThreadCas alias (CAS is global, not per-thread)
- Remove formatSkillDoc() legacy compat shim
- Remove help command (replaced by skill)
- Remove all 13 DEPRECATED_ALIASES flat commands + printDeprecation
- Fix CAS prompts in develop roles: remove stale <THREAD_ID> param
- Update README.md to remove createThreadCas reference
- Net: -86 lines, 241 tests pass

Closes #114
2026-05-08 10:27:27 +08:00
xiaomo 085cdcd3f4 Merge pull request 'feat: engine injects extract provider at runtime (Phase 2)' (#113) from feat/110-phase2-migrate-extract into main 2026-05-08 02:23:58 +00:00
xiaoju a8c1c158d6 feat: engine injects extract provider at runtime (Phase 2)
- createWorkflow(def, binding) — no more extract/llmProvider params
- Engine resolves extract provider from workflow.yaml via resolveModel
- WorkflowFnOptions now carries extract + llmProvider (engine-injected)
- Delete extract-provider.ts, inline maxDepth helper
- Template packages simplified: only take agent binding
- Breaking change: bundles no longer carry provider config

Refs #110
2026-05-08 02:21:45 +00:00
xiaomo 83649fd836 Merge pull request 'docs: add README.md to all 8 packages' (#112) from docs/package-readmes into main 2026-05-08 02:19:25 +00:00
xingyue a5c09adae6 docs: add README.md to all 8 packages
Each README includes: package description, install instructions,
usage example, API overview, and (for templates) role/moderator flow.

Packages documented:
- @uncaged/workflow (core)
- @uncaged/cli-workflow (CLI)
- @uncaged/workflow-agent-cursor
- @uncaged/workflow-agent-hermes
- @uncaged/workflow-agent-llm
- @uncaged/workflow-template-develop
- @uncaged/workflow-template-solve-issue
- @uncaged/workflow-util-agent
2026-05-08 10:17:40 +08:00
xiaoju 1f1128ff4a fix: address PR #111 review feedback
- Extract validateWorkspaceSegment to commands/init/validate.ts
- Unify splitProviderModelRef in config/, used by both resolve-model and registry-normalize
- Warn on missing models.default during parse (tag Z2KP9NWQ)
2026-05-08 02:14:20 +00:00
xiaoju aa01283ce1 feat: unified provider/model configuration (Phase 1)
- New src/config/ folder: resolveModel(config, scene) with fallback to default
- WorkflowConfig now has providers + models instead of extract
- Delete ExtractProviderConfig, getExtractProvider uses resolveModel('extract')
- New resolve-model tests, updated existing tests

Refs #110
2026-05-08 02:08:19 +00:00
xiaoju 2bbe5a3d0e chore: enforce folder module discipline in @uncaged/workflow
Each folder now has:
- types.ts for all type definitions
- index.ts with pure re-exports only
- Cross-folder imports go through index.ts

Closes #106
2026-05-08 01:37:23 +00:00
xiaoju cf17dedac3 refactor: organize workflow/src into 6 module folders
Move 34 flat modules into cas/, registry/, bundle/, extract/, engine/, util/.
Move gc.ts to engine/ (was in cas/) to avoid cas→engine reverse dependency.
Dependency direction: util ← cas ← extract ← engine, util ← registry ← bundle.
No logic changes — only file locations and import paths.

Refs #102
2026-05-08 01:22:01 +00:00
xiaoju 485bfcb0b6 chore: remove unused build scripts
All packages are pure Bun/TS — no build step needed.
The build scripts were all placeholder `echo 'TODO'` anyway.
2026-05-08 00:46:35 +00:00
xiaoju 71b413f20c feat(planner): add phase granularity guidance to reduce over-splitting
Simple tasks were getting 3 phases when 1 would suffice. Added explicit
complexity-to-phase-count mapping in the planner system prompt.

小橘 🍊
2026-05-07 14:20:37 +00:00
xingyue 8fe26417cf feat(cli): add --latest, --debug, --role flags to live command (#37 Phase 2)
- --latest: auto-find most recent thread by start timestamp
- --debug: display .info.jsonl debug log with tags
- --role: filter output to specific role
- Add live-argv.ts for flag parsing
- Add fixtures and test coverage for all flags

Testing: #50
2026-05-07 21:44:19 +08:00
xingyue 990200230b feat(cli): add live command for real-time thread monitoring (#37 Phase 1)
- Add cmd-live.ts: tail .data.jsonl with formatted output
- Display role steps with timestamp, role name, truncated content, meta
- fs.watch for running threads, auto-exit on completion
- Write WorkflowResult to .data.jsonl in worker.ts for completion detection
- Add live.test.ts with JSONL fixtures

Testing: #49
2026-05-07 21:42:32 +08:00
xiaoju 43a6600378 feat: ReAct ExtractFn with tool-use
- RoleDefinition.extractMode: "single" | "react"
- reactExtract: multi-turn LLM with cas_get tool for DAG traversal
- Max 10 tool-call rounds, schema validation on final output
- create-workflow routes to reactExtract when extractMode is "react"
- All existing roles set to "single" (no behavior change)
- 162 tests passing

Fixes #44
2026-05-07 13:28:00 +00:00
xiaoju cae59b589e feat: global extract provider config
- workflow.yaml supports config section (maxDepth, extract provider)
- ExtractProviderConfig with env: prefix for apiKey resolution
- getExtractProvider(storageRoot) returns LlmProvider from config
- workflowAsAgent uses config maxDepth (fallback 3)
- Registry read/write preserves config
- 158 tests passing

Fixes #43
2026-05-07 13:21:38 +00:00
xiaoju 6196e0974a feat: thread root node + workflowAsAgent returns root hash
- Engine generates step Merkle nodes (type: step) after each role step
- Engine generates thread root Merkle node (type: thread) on completion
- WorkflowResult gains rootHash field
- WorkflowFn returns WorkflowCompletion (no rootHash), engine wraps with rootHash
- workflowAsAgent returns rootHash instead of summary
- Full DAG traversal: root → steps → content
- 151 tests passing

Fixes #42
2026-05-07 13:17:44 +00:00
xiaoju 84de74721d feat: Merkle node format + content → CAS
- MerkleNode type: { type, payload, children } serialized as YAML
- RoleOutput.content → contentHash (CAS hash of Merkle content node)
- Engine stores content in global CAS before writing to .data.jsonl
- create-workflow puts content as Merkle node, merges contentHash into refs
- fork/parse adapted for contentHash format
- buildAgentPrompt now async, reads content from CAS
- Bundle validator allows @uncaged/workflow import
- 150 tests passing

BREAKING: .data.jsonl no longer contains inline content

Fixes #41
2026-05-07 13:14:01 +00:00
xiaoju e95e76c145 feat: workflowAsAgent factory
- workflowAsAgent(name) resolves via registry → bundle → child thread
- System-level depth limit (max 3, constant)
- Returns summary string, errors as string (no throw)
- Integration test with nested workflow execution
- 146 tests passing

Fixes #33
2026-05-07 10:52:26 +00:00
xiaoju 6488b7bbb4 feat: CAS garbage collection
- garbageCollectCas() mark-and-sweep: scan .data.jsonl refs, delete orphans
- 'uncaged-workflow gc' CLI command
- thread rm triggers GC automatically
- 141 tests passing

Fixes #32
2026-05-07 10:47:52 +00:00
xiaoju 30e4e99908 feat: add refs tracking to RoleStep
- RoleOutput gains refs: string[] for CAS reference tracking
- RoleDefinition gains extractRefs: ((meta) => string[]) | null
- planner: phases.map(p => p.hash), coder: [completedPhase]
- Engine persists refs, fork preserves refs
- Backward compat: missing refs normalized to []
- 137 tests passing

Fixes #31
2026-05-07 10:44:25 +00:00
xiaoju 12d58a8206 feat: migrate CAS to global storage
- Add getGlobalCasDir() to storage-root.ts
- cmd-cas.ts uses global CAS dir, threadId kept for CLI compat
- thread rm no longer deletes .cas/ directories
- Rename createThreadCas → createCasStore (deprecated alias kept)
- 134 tests passing

BREAKING: CAS moves from <thread>.cas/ to <storageRoot>/cas/

Fixes #30
2026-05-07 10:40:14 +00:00
xiaoju f73bf1e313 test(workflow): add unit tests for validateWorkflowDescriptor
Fixes #19
2026-05-07 09:38:18 +00:00
xiaoju 4b44665c7e feat(workflow): add thread-scoped CAS (Content-Addressable Storage)
Phase 1 of #23:
- createThreadCas() core API: put/get/delete/list with XXH64 hashing
- hashString() utility for string → 13-char Crockford Base32
- CLI: uncaged-workflow cas get/put/list/rm subcommands
- thread rm now cleans up .cas/ directory
- 10 new tests for CAS operations

Refs #23
2026-05-07 04:30:19 +00:00
xiaoju 43e1f82303 refactor: extractPrompt out of ExtractContext, into ExtractFn parameter
ExtractFn = (schema, prompt, ctx) => Promise<T>
extractPrompt stays in RoleDefinition (definition layer), not in context (state layer).
Callers pass their own prompt — engine uses roleDef.extractPrompt, cursor agent uses its own.

小橘 <xiaoju@shazhou.work>
2026-05-07 01:27:12 +00:00
xiaoju d472de1247 refactor: three-phase context (Moderator/Agent/Extract) + extractPrompt + unified ExtractFn
- ModeratorContext → AgentContext → ExtractContext progressive types
- ThreadContext is now alias for AgentContext
- RoleDefinition adds extractPrompt field
- ExtractFn = (schema, ctx: ExtractContext) => Promise<T>
- createWorkflow takes ExtractFn, engine loop: moderator → agent → extract
- Remove ExtractConfig, extractMetaOrThrow, extract-meta.ts

小橘 <xiaoju@shazhou.work>
2026-05-07 01:05:31 +00:00
xiaoju 99a137422c feat: add ExtractFn utility, cursor agent workspace from thread context
- New ExtractFn = <T>(schema, prompt) => (ctx) => Promise<T>
- createExtract(provider) creates an LLM-backed ExtractFn
- CursorAgent removes workdir config, uses ExtractFn to resolve workspace from ThreadContext at runtime
- buildAgentPrompt(ctx) — reads systemPrompt from ctx.currentRole

小橘 <xiaoju@shazhou.work>
2026-05-07 00:17:31 +00:00
xiaoju 2482fb7e62 chore: remove all dryRun infrastructure
dryRun no longer needed — tests use mock agents + mock fetch instead.
Removes isDryRun from WorkflowFnOptions, dryRun from ExtractConfig,
dryRunMeta from RoleDefinition, --dry-run from CLI, and all related
plumbing in engine/worker/fork/extract.

小橘 <xiaoju@shazhou.work>
2026-05-06 14:25:44 +00:00
xiaoju fa9163e462 refactor: all-agentic architecture — roles as pure data, agent binding at runtime
BREAKING: Major architecture change.

- RoleDefinition = pure data (systemPrompt + schema + dryRunMeta)
- AgentFn = (ctx: ThreadContext) => Promise<string>, reads ctx.currentRole
- WorkflowDefinition decoupled from agents, bound via AgentBinding at runtime
- createWorkflow(def, binding, extract) replaces createRoleModerator
- Meta extraction moved into engine loop
- Delete workflow-util-role package (createRole, decorators, extract all gone)
- Role packages become pure data exports
- Agent packages updated to single-arg AgentFn

小橘 <xiaoju@shazhou.work>
2026-05-06 14:14:33 +00:00
xiaoju c7b0beb6be refactor: unify RoleDefinition + WorkflowDefinition with description & schema
- Add RoleDefinition<Meta> = { description, run, schema } to core types
- WorkflowDefinition now carries description and RoleDefinition per role
- Add buildDescriptor() in core to derive WorkflowDescriptor from WorkflowDefinition
- Remove buildDescriptorFromRoles / RoleDescriptorInput from workflow-util-role
- Update solve-issue template, examples, and all tests

小橘 <xiaoju@shazhou.work>
2026-05-06 11:19:49 +00:00
xiaoju 79cf97e617 refactor: remove name from WorkflowDefinition, fix threadId type errors
WorkflowDefinition no longer carries 'name' — name is a registry-level
concern, not a bundle property. Also fixed all missing threadId in test
fixtures to match the updated ThreadContext type.

小橘 <xiaoju@shazhou.work>
2026-05-06 11:01:09 +00:00
xiaoju 8d5b97c67e feat: add threadId to ThreadContext and WorkflowFnOptions
Agentic roles can now access ctx.threadId to query thread details
via uncaged-workflow CLI for richer context.
2026-05-06 10:23:25 +00:00
xiaoju c15a5554c0 refactor: replace gitExec with spawnCli from workflow-util-agent
Remove hand-rolled execFile + promisify, delegate to spawnCli.
Same throw-on-error interface, better error messages.
2026-05-06 09:51:06 +00:00
xingyue 98b6153070 fix(workflow): resolve type errors across all packages and remove tsbuildinfo from tracking
- Add bun-types and @types/xxhashjs to root devDependencies
- Add types: ['bun-types'] to root and package tsconfigs
- Fix AST Node type narrowing in bundle-validator.ts with AcornNode type and narrowNode helper
- Fix generic ThreadContext variance in create-role-moderator.ts
- Add explicit parameter types in worker.ts
- Fix ChildProcess type in worker-spawn.ts to match spawn() stdio config
- Remove all tsconfig.tsbuildinfo from git tracking
- Add tsconfig.tsbuildinfo to .gitignore
2026-05-06 17:41:11 +08:00
xiaoju f21014fcdd feat: @uncaged/workflow-agent-cursor + @uncaged/workflow-agent-hermes
- Cursor adapter: spawn cursor-agent CLI, auto/specified model
- Hermes adapter: spawn hermes chat CLI
- Both: AgentFn interface, no nerve-core deps, Result-based config validation
- 93 tests pass, biome clean

Closes #10, Closes #11
小橘 <xiaoju@shazhou.work>
2026-05-06 06:54:24 +00:00
xiaoju 3467b772e6 refactor: named exports (run + descriptor), remove build pipeline
- Bundle contract: export const run + export const descriptor (no default export)
- add only accepts .esm.js, extracts descriptor via dynamic import → .yaml
- Removed: build-pipeline, generate-types, json-schema-to-ts
- Worker loads mod.run instead of mod.default
- Biome: no more noDefaultExport overrides for bundles
- 62 tests pass, biome clean

Closes #8
小橘 <xiaoju@shazhou.work>
2026-05-06 06:39:15 +00:00
xiaoju e670047e6a feat: build pipeline — .ts → .esm.js + .yaml + .d.ts 三件套
- add command auto-detects .ts vs .esm.js input
- .ts: Bun.build → bundle + descriptor extraction + JSON Schema → .d.ts
- .esm.js: requires .yaml alongside, .d.ts optional
- JSON Schema → TypeScript type converter
- hello-world example workflow
- 63 tests pass, biome clean

Closes #7
小橘 <xiaoju@shazhou.work>
2026-05-06 06:26:14 +00:00
xiaoju 47e8fdf5b3 chore: replace hand-written xxhashjs.d.ts with @types/xxhashjs
小橘 <xiaoju@shazhou.work>
2026-05-06 06:21:46 +00:00
xiaoju dfbba0f58c feat: Phase 4 — fork threads + bun publish verified
- fork-thread.ts: parse .data.jsonl, trim steps by role
- cmd-fork.ts: --from-role <role> or retry last step
- engine: forkFrom lineage tracking, prefilled step replay
- worker: accept steps in run IPC command
- bun publish --dry-run: both packages pass
- 53 tests pass, biome clean

Closes #5
小橘 <xiaoju@shazhou.work>
2026-05-06 05:45:01 +00:00
xiaoju 0becafeb44 feat: Phase 3 — version history/rollback + pause/resume threads
- CLI: history, rollback, pause, resume commands
- Registry: rollbackWorkflowToHistoryHash
- Engine: awaitAfterEachYield hook for pause gate
- Worker: ThreadPauseGate with Promise-based latch
- TCP IPC: bidirectional response for kill/pause/resume
- 44 tests pass, biome clean

小橘 <xiaoju@shazhou.work>
2026-05-06 05:36:33 +00:00
xiaoju 9943f21f5c refactor: WorkflowFn input → ThreadInput, remove threadId from bundle contract
- WorkflowFn first param is now ThreadInput { prompt, steps }
- threadId removed from WorkflowFnOptions and ThreadContext (engine-only)
- createRoleModerator seeds context from input.steps (fork/resume ready)
- New test: pre-filled steps skip already-completed roles

Closes #6
小橘 <xiaoju@shazhou.work>
2026-05-06 05:27:14 +00:00
xiaoju 9a4cec2b2d docs(rfc-001): WorkflowFn input → ThreadInput for fork/resume support
- First param is now { prompt, steps } instead of bare prompt
- steps: [] for new thread, pre-filled for fork/resume
- createRoleModerator naturally handles resume via moderator routing
- No special replay logic needed

小橘 <xiaoju@shazhou.work>
2026-05-06 05:25:00 +00:00
xiaoju 7582a88d6b feat: Phase 2 — Thread lifecycle, execution engine, worker, CLI
- types.ts: START/END, RoleMeta, ThreadContext, Role, Moderator, WorkflowDefinition
- engine.ts: executeThread with JSONL persistence + AbortSignal
- worker.ts: per-bundle process, TCP IPC, kill individual threads
- CLI: run/ps/kill/threads/thread/thread rm commands
- 32 tests pass, biome clean

小橘 <xiaoju@shazhou.work>
2026-05-06 04:59:54 +00:00
xiaoju 01e930df8f docs(rfc-001): add execution model — Role, Moderator, Agent types
Ported from nerve's workflow types. Covers ThreadContext, StartStep,
RoleStep, Moderator (pure router), Role (async actor), AgentFn (LLM adapter),
WorkflowDefinition, and execution flow.

小橘 <xiaoju@shazhou.work>
2026-05-06 04:41:52 +00:00
xiaoju 8939194133 init: bun workspace + RFC-001 workflow engine design
- @uncaged/workflow (core lib) + @uncaged/cli-workflow (CLI)
- RFC-001: full design doc covering storage, threading, CLI

小橘 <xiaoju@shazhou.work>
2026-05-06 04:20:05 +00:00