analyzing-codebases
Detects project languages and monorepo state, runs language-appropriate static analysis (dependency graph, complexity, duplication, semantic patterns), and…
Writes AGENTS.md per subproject, archives run artifacts, and suggests rcc handoff conditionally. Use when verifying-refactors passes (PASS or PASS-WITH-WEAK-TESTS).
$ npx -y skills add wayne930242/Reflexive-Claude-Code --skill finalizing-refactors --agent claude-codeHow it fires
How this skill gets triggered: by you, by Claude, or both.
/finalizing-refactorsContext preview
The summary Claude sees to decide when to auto-load this skill.
Writes AGENTS.md per subproject, archives run artifacts, and suggests rcc handoff conditionally. Use when verifying-refactors passes (PASS or PASS-WITH-WEAK-TESTS).
name: finalizing-refactors description: Writes AGENTS.md per subproject, archives run artifacts, and suggests rcc handoff conditionally. Use when verifying-refactors passes (PASS or PASS-WITH-WEAK-TESTS).
**Finalizing refactors IS leaving a codebase that the next AI agent can pick up immediately.**
Write per-subproject AGENTS.md (the cross-tool standard for agent-targeted docs), archive run artifacts to `.rcc/archive/`, suggest next-step handoff to rcc based on project state.
**Core principle:** The refactor is not done until the next agent has a map.
**Pattern:** Terminal **Handoff:** user-confirmation **Next:** none (terminal), but suggests `/rcc migrate` or `/rcc reflect`
**Tasks:** 1. Detect subprojects 2. For each subproject, read existing AGENTS.md (if any) 3. Generate AGENTS.md content per template 4. Show diff and get user approval 5. Write AGENTS.md files 6. Archive run artifacts 7. Detect rcc state and suggest handoff 8. Print final summary
Root has a manifest → root is a subproject. Any directory with its own manifest and not nested in another is a subproject. Use same detection as analyzing-codebases.
For each subproject root, check if `AGENTS.md` exists. If yes, parse existing section headings. Plan will add missing sections, preserve existing.
Per subproject, fill template at `references/agents-md-template.md`:
Show `diff -u <existing> <proposed>` per subproject. User approves per subproject or batch. Rejections → prompt for edits.
Write approved AGENTS.md files. Commit:
git add <paths> git commit -m "docs(aref): add/update AGENTS.md for N subprojects"
Move run artifacts to `.rcc/archive/{ts}-aref-run/`:
mkdir -p .rcc/archive/{ts}-aref-run
mv .rcc/{ts}-*.md .rcc/archive/{ts}-aref-run/
mv .rcc/{ts}-*.json .rcc/archive/{ts}-aref-run/ 2>/dev/null || true
mv .rcc/{ts}-apply-state.yml .rcc/archive/{ts}-aref-run/ 2>/dev/null || trueKeep `.rcc/aref-raw/` and clean it: `rm -rf .rcc/aref-raw/{ts}-*`.
Commit:
git add .rcc/archive/{ts}-aref-run/
git commit -m "chore(aref): archive run {ts}"Check:
No agent system detected → print: > Refactor complete. This project has no agent system. Next: `/rcc migrate` to bootstrap skills, CLAUDE.md, rules.
Agent system present → print: > Refactor complete. Existing agent system detected. Next: `/rcc reflect` to let the existing skills incorporate the new structure.
Print:
Instruct user: > To merge: `git checkout main && git merge --no-ff <branch>` or open a PR.
| Thought | Reality | |---------|---------| | "AGENTS.md is redundant with README" | AGENTS.md is the agent-targeted doc; README is human-targeted. Different audiences, different content. | | "Skip archive, user can read .rcc/ directly" | Archive keeps unstacked run history. Cleans working .rcc/. | | "Merge the branch to finalize" | User merges. Plugin stops at branch. |
A Claude Code plugin marketplace for skills-driven Agentic Context Engineering (ACE) — build, analyze, and maintain agent systems with structured workflows.
Repo: wayne930242/Reflexive-Claude-Code
Detects project languages and monorepo state, runs language-appropriate static analysis (dependency graph, complexity, duplication, semantic patterns), and…
Executes a refactor plan phase-by-phase on a dedicated branch with per-phase commits and mandatory reviewer checkpoints. Use when characterization-tests…
Converts a refactor map into a phased plan using parallel-change, branch-by-abstraction, or strangler fig patterns. Use when user has approved the refactor map…
Adds golden/snapshot tests to untested hotspot modules before refactoring. Use when refactor plan marks any phase with…
Validates hard structural rules (no cycles, file/fn line caps, cognitive/cyclomatic complexity) and runs mutation testing on touched modules. Use when…
Validates component-type choices for agent system work, classifying knowledge as CLAUDE.md vs rule vs skill vs agent vs hook and checking for conflicts. Use…