- Rewrite supervisor to use createThreadReactor + createLlmFn
- No direct fetch/HTTP calls in supervisor
- All 266 tests passing
Refs #139, relates #141
Redefine WorkflowFn to accept a complete ThreadContext plus WorkflowRuntime dependencies, removing ThreadInput and WorkflowFnOptions.
Move thread context construction into engine executeThread, update runtime loop/agent paths, and align templates/docs/tests with template-only definition exports.
Co-authored-by: Cursor <cursoragent@cursor.com>
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
- 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
- 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
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>
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>
- 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>
- 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>
xiaoju
Scott Wei