Files
Claude-Code-Workflow/.codex/skills/team-lifecycle-v4/specs/knowledge-transfer.md
catlog22 1e560ab8e8 feat: migrate all codex team skills from spawn_agents_on_csv to spawn_agent + wait_agent architecture
- Delete 21 old team skill directories using CSV-wave pipeline pattern (~100+ files)
- Delete old team-lifecycle (v3) and team-planex-v2
- Create generic team-worker.toml and team-supervisor.toml (replacing tlv4-specific TOMLs)
- Convert 19 team skills from Claude Code format (Agent/SendMessage/TaskCreate)
  to Codex format (spawn_agent/wait_agent/tasks.json/request_user_input)
- Update team-lifecycle-v4 to use generic agent types (team_worker/team_supervisor)
- Convert all coordinator role files: dispatch.md, monitor.md, role.md
- Convert all worker role files: remove run_in_background, fix Bash syntax
- Convert all specs/pipelines.md references
- Final state: 20 team skills, 217 .md files, zero Claude Code API residuals

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-03-24 16:54:48 +08:00

4.8 KiB

Knowledge Transfer Protocols

1. Transfer Channels

Channel Method Producer Consumer
Artifacts Files in <session>/artifacts/ Task executor Next task in pipeline
Discoveries Files in <session>/discoveries/{task_id}.json Task executor Coordinator + downstream
Wisdom Append to <session>/wisdom/*.md Any role All roles
Context Accumulator In-memory aggregation Inner loop only Current task
Exploration Cache <session>/explorations/ Analyst / researcher All roles

2. Context Loading Protocol (Before Task Execution)

Every role MUST load context in this order before starting work.

Step Action Required
1 Read <session>/tasks.json -- locate upstream task entries, check status and findings Yes
2 Read <session>/discoveries/{upstream_id}.json for each upstream dependency -- get detailed findings and artifact paths Yes
3 Read artifact files from upstream discovery's artifacts_produced paths Yes
4 Read <session>/wisdom/*.md if exists Yes
5 Check <session>/explorations/cache-index.json before new exploration If exploring

Loading rules:

  • Never skip step 1 -- tasks.json contains task status, wave progression, and summary findings
  • Never skip step 2 -- discoveries contain detailed key_findings, decisions, and artifact references
  • If artifact path in discovery does not exist, log warning and continue
  • Wisdom files are append-only -- read all entries, newest last

3. Context Publishing Protocol (After Task Completion)

Step Action Required
1 Write deliverable to <session>/artifacts/<task-id>-<name>.md Yes
2 Write <session>/discoveries/{task_id}.json with payload (see schema below) Yes
3 Append wisdom entries for learnings, decisions, issues found If applicable

4. Discovery File Schema

Written to <session>/discoveries/{task_id}.json on task completion.

{
  "task_id": "<TASK-NNN>",
  "worker": "<TASK-NNN>",
  "timestamp": "2026-03-24T10:15:00+08:00",
  "type": "<pipeline_phase>",
  "status": "completed | failed",
  "findings": "Summary string (max 500 chars)",
  "quality_score": null,
  "supervision_verdict": null,
  "error": null,
  "data": {
    "key_findings": [
      "Finding 1",
      "Finding 2"
    ],
    "decisions": [
      "Decision with rationale"
    ],
    "files_modified": [
      "path/to/file.ts"
    ],
    "verification": "self-validated | peer-reviewed | tested"
  },
  "artifacts_produced": [
    "<session>/artifacts/<task-id>-<name>.md"
  ]
}

Field rules:

  • artifacts_produced: Always artifact paths, never inline content
  • data.key_findings: Max 5 items, each under 100 chars
  • data.decisions: Include rationale, not just the choice
  • data.files_modified: Only for implementation tasks
  • data.verification: One of self-validated, peer-reviewed, tested

Supervisor-specific extensions (CHECKPOINT tasks only):

{
  "supervision_verdict": "pass | warn | block",
  "supervision_score": 0.85,
  "data": {
    "risks_logged": 0,
    "blocks_detected": 0
  }
}
  • supervision_verdict: Required for CHECKPOINT tasks. Determines pipeline progression.
  • supervision_score: Float 0.0-1.0. Aggregate of individual check scores.
  • data.risks_logged: Count of risks written to wisdom/issues.md.
  • data.blocks_detected: Count of blocking issues found. >0 implies verdict=block.

5. Exploration Cache Protocol

Prevents redundant research across tasks.

Step Action
1 Read <session>/explorations/cache-index.json
2 If angle already explored, read cached result from explore-<angle>.json
3 If not cached, perform exploration
4 Write result to <session>/explorations/explore-<angle>.json
5 Update cache-index.json with new entry

cache-index.json format:

{
  "entries": [
    {
      "angle": "competitor-analysis",
      "file": "explore-competitor-analysis.json",
      "created_by": "RESEARCH-001",
      "timestamp": "2026-01-15T10:30:00Z"
    }
  ]
}

Rules:

  • Cache key is the exploration angle (normalized to kebab-case)
  • Cache entries never expire within a session
  • Any role can read cached explorations; only the creator updates them

6. Platform Mapping Reference

Claude Code Operation Codex Equivalent
team_msg(operation="get_state", role=<upstream>) Read tasks.json + Read discoveries/{upstream_id}.json
team_msg(type="state_update", payload={...}) Write discoveries/{task_id}.json
TaskCreate / TaskUpdate status fields Read/Write tasks.json task entries
In-memory state aggregation Parse tasks.json + glob discoveries/*.json