mirror of https://github.com/catlog22/Claude-Code-Workflow.git synced 2026-03-25 19:48:33 +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

3.3 KiB

Raw Blame History

role, prefix, inner_loop, message_types

role

prefix

inner_loop

message_types

fixer

FIX

true

success	error
fix_complete	fix_failed

Code Fixer

Fix code based on reviewed findings. Load manifest, plan fix groups, apply with rollback-on-failure, verify. Code-generation role -- modifies source files.

Phase 2: Context & Scope Resolution

Input	Source	Required
Task description	From task subject/description	Yes
Session path	Extracted from task description	Yes
Fix manifest	/fix/fix-manifest.json	Yes
Review report	/review/review-report.json	Yes
.msg/meta.json	/.msg/meta.json	No

Extract session path, input path from task description
Load manifest (scope, source report path) and review report (findings with enrichment)
Filter fixable findings: severity in scope AND fix_strategy !== 'skip'
If 0 fixable -> report complete immediately
Detect quick path: findings <= 5 AND no cross-file dependencies
Detect verification tools: tsc (tsconfig.json), eslint (package.json), jest (package.json), pytest (pyproject.toml), semgrep (semgrep available)
Load wisdom files from <session>/wisdom/

Phase 3: Plan + Execute

3A: Plan Fixes (deterministic, no CLI)

Group findings by primary file
Merge groups with cross-file dependencies (union-find)
Topological sort within each group (respect fix_dependencies, append cycles at end)
Sort groups by max severity (critical first)
Determine execution path: quick_path (<=5 findings, <=1 group) or standard
Write <session>/fix/fix-plan.json: {plan_id, quick_path, groups[{id, files[], findings[], max_severity}], execution_order[], total_findings, total_groups}

3B: Execute Fixes

Quick path: Single code-developer agent for all findings. Standard path: One code-developer agent per group, in execution_order.

Agent prompt includes: finding list (dependency-sorted), file contents (truncated 8K), critical rules:

Apply each fix using Edit tool in order
After each fix, run related tests
Tests PASS -> finding is "fixed"
Tests FAIL -> git checkout -- {file} -> mark "failed" -> continue
No retry on failure. Rollback and move on
If finding depends on previously failed finding -> mark "skipped"

Agent returns JSON: {results:[{id, status: fixed|failed|skipped, file, error?}]} Fallback: check git diff per file if no structured output.

Write <session>/fix/execution-results.json: {fixed[], failed[], skipped[]}

Phase 4: Post-Fix Verification

Run available verification tools on modified files:

Tool	Command	Pass Criteria
tsc	`npx tsc --noEmit`	0 errors
eslint	`npx eslint <files>`	0 errors
jest	`npx jest --passWithNoTests`	Tests pass
pytest	`pytest --tb=short`	Tests pass
semgrep	`semgrep --config auto <files> --json`	0 results

If verification fails critically -> rollback last batch
Write <session>/fix/verify-results.json
Generate <session>/fix/fix-summary.json: {fix_id, fix_date, scope, total, fixed, failed, skipped, fix_rate, verification}
Generate <session>/fix/fix-summary.md (human-readable)
Update <session>/.msg/meta.json with fix results
Contribute discoveries to <session>/wisdom/ files

3.3 KiB Raw Blame History