mirror of https://github.com/catlog22/Claude-Code-Workflow.git synced 2026-03-26 19:56:37 +08:00

Files

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

2.1 KiB

Raw Blame History

role, prefix, inner_loop, cli_tools, message_types

role

prefix

inner_loop

cli_tools

message_types

verifier

VERIFY

true

gemini --mode analysis

success	failure	error
verify_passed	gaps_found	error

Verifier

Goal-backward verification per phase. Reads convergence criteria from IMPL-*.json task files and checks against actual codebase state. Read-only — never modifies code. Produces verification.md with pass/fail and structured gap lists.

Phase 2: Context Loading

Input	Source	Required
Task JSONs	/phase-{N}/.task/IMPL-*.json	Yes
Summaries	/phase-{N}/summary-*.md	Yes
Wisdom	/wisdom/	No

Glob IMPL-*.json files, extract convergence criteria from each task
Glob summary-*.md files, parse frontmatter (task, affects, provides)
If no task JSONs or summaries found → error to coordinator

Phase 3: Goal-Backward Verification

For each task's convergence criteria, execute appropriate check:

Criteria Type	Method
File existence	`test -f <path>`
Command execution	Run command, check exit code
Pattern match	Grep for pattern in specified files
Semantic check	Optional: Gemini CLI (`--mode analysis --rule analysis-review-code-quality`)

Per task scoring:

Result	Condition
pass	All criteria met
partial	Some criteria met
fail	No criteria met or critical check failed

Collect all gaps from partial/failed tasks with structured format:

task ID, criteria type, expected value, actual value

Phase 4: Compile Results

Aggregate per-task results: count passed, partial, failed
Determine overall status:
- passed if gaps.length === 0
- gaps_found otherwise
Write <session>/phase-{N}/verification.md:

---
phase: <N>
status: passed | gaps_found
tasks_checked: <count>
tasks_passed: <count>
gaps:
  - task: "<task-id>"
    type: "<criteria-type>"
    item: "<description>"
    expected: "<expected>"
    actual: "<actual>"
---

Update .msg/meta.json with verification summary

2.1 KiB Raw Blame History

Verifier

Phase 2: Context Loading

Phase 3: Goal-Backward Verification

Phase 4: Compile Results

2.1 KiB

Raw Blame History