safe-refactor
Skill0 starsChanged 18 days agoMITView on GitHub
Behavior-preserving refactoring across languages and stacks: plan multi-file work, execute one safe boundary at a time, review-and-align to project standards, cut method complexity, apply smell/pattern catalogs, and run scoped tracks for Go, Elixir (via super-elixir), Terraform modules, React components, web UI, and mobile placeholders. Always pairs with output-savers for minimal diffs. Use when the user asks to refactor, clean up, extract, rename, reduce complexity, split a component, refactor a Terraform module, plan a safe refactor, or fix UI hierarchy — not for new features or repo rebuilds.
26 other projects use a version of this.
Adopted by
Most recently taken up by
- sudeepgupta90/my-ai-marketplacetheir own version, first seen Nov 8, 2026
- mjlueck-hash/cavemantheir own version, first seen Nov 8, 2026
- dhruvsaxena1998/skillstheir own version, first seen Nov 8, 2026
- MMACABRE/SKILLStheir own version, first seen Nov 8, 2026
- Terryc21/sitreptheir own version, first seen Nov 1, 2026
- agentkitco/agentkit-dev-skillstheir own version, first seen Nov 1, 2026
Used with
Projects with it also use verify-and-stop, investigate-first and 10 more.
Other files people use alongside this one.
Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
JuliusBrusseeSkillverify-and-stop
Prove existing work meets acceptance conditions without expanding scope. Use for validation-only tasks, completion checks, focused gate runs, and last-mile proof.
Also in 8 projects that use this
- JuliusBrusseeSkill
investigate-first
Diagnose ambiguous failures before editing. Use for unknown causes, intermittent behavior, performance regressions, or investigations needing evidence-ranked hypotheses.
Also in 7 projects that use this
- +10 more from JuliusBrussee
Other versions26 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 26
- JuliusBrussee/caveman★ 118ksame name
- Terryc21/sitrep★ 50same name
- JPeetz/agent-skills★ 14same name
- yeaight7/agent-powerups★ 6same name
- dfedoryshchev/weftworks★ 5same name
- Linchi1997yeh/craftsman-suite★ 2same name
- nomdedev/agent-manager-template★ 2same name
- xingtu1996/dsh-xingtu-skills★ 2same name
- dennisrongo/claude-skills★ 1same name
- dhruvsaxena1998/skills★ 1same name
16 more not listed
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository57 other files it gives its agents.
- AGENTS.mdAGENTS.md
- .cursor/rules/comments.mdcCursor rule
- .cursor/rules/commit-messages.mdcCursor rule
- .cursor/rules/git-workflow.mdcCursor rule
- .cursor/rules/manage-stack.mdcCursor rule
- .cursor/rules/no-worktrees.mdcCursor rule
- manage-stackSkill
- convert-as-skillSkill
- create-agentSkill
- create-agents-mdSkill
- create-commandSkill
- create-hookSkill
- create-ruleSkill
- create-skillSkill
- retrospectiveSkill
- super-architectureSkill
- super-cachingSkill
- super-elixirSkill
- super-nodejsSkill
- super-protobufSkill
- super-tech-blueprintSkill
- atomic-semantic-commitSkill
- code-reviewSkill
- create-bruno-collectionSkill
- create-documentationSkill
- create-ticketsSkill
- ready-specSkill
- write-adrSkill
- write-changelogSkill
- write-mr-descriptionSkill
- write-planSkill
- write-postmortemSkill
- write-runbookSkill
- noslopSkill
- output-saversSkill
- tddSkill
- verifySkill
- get-movingSkill
- later-listSkill
- standupSkill
- challenge-meSkill
- codebase-learnerSkill
- researchSkill
- second-opinionSkill
- traceSkill
- color-kitSkill
- design-system-kitSkill
- diagramSkill
- infographicSkill
- product-polish-workflowSkill
- prototype-wireframeSkill
- ui-ux-kitSkill
- evidenceSkill
- find-skillsSkill
- handoffSkill
- rumblingSkill
- use-levistackSkill
The file
We split this skill into the rules it gives an agent
2 rules. 0 are common, 2 are distinctive, and 0 are very specific to this project.
0
Common
Most of the 24 edited versions we read keep these.
2
Distinctive
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 33
Simplicity: Before editing code, apply output-savers via references/core/output-savers-integration.md.
- Distinctive: line 39
Do this first; do not announce it.
“Common” is measured: kept by at least 6 in 10 of the 24 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Safe refactor
One router for behavior-preserving structural improvement. Gradual evolution, not rewrite. Feature work stays out of scope unless the user explicitly combines them.
Simplicity: Before editing code, apply output-savers via references/core/output-savers-integration.md. Missing tests on the path you will move: load tdd first, then return here.
Research notes: references/core/research-synthesis.md. Stack index: references/tech/README.md.
Step 0: detect the job
Do this first; do not announce it. Full signal table: references/core/auto-detect.md.
| If the user is doing this | Mode | Load |
| Plan, sequence, or scope multi-file refactor | plan | refactor-plan-protocol.md |
| Execute extract/rename/move/cleanup with tests | safe-execute | safe-refactor-protocol.md |
| Review against repo rules then refactor | review-and-align | review-and-align-protocol.md |
| Cut cognitive complexity of one method | reduce-complexity | reduce-complexity-protocol.md |
| Smells, duplication, Fowler patterns | patterns/smells | refactoring-patterns-overview.md, legacy-code-smells.md |
| Textual / structural / conceptual dupes | duplication-levels | duplication-levels.md |
| Go files, gopls, import cycles | language: Go | refactoring.md |
| Elixir / OTP / Ecto idioms | language: Elixir | elixir.md → super-elixir |
| Terraform module extraction | language: Terraform | module-refactor.md |
| React/TSX split, huge components | language: React | component-refactor.md |
| Visual hierarchy, spacing, Tailwind | ui-refactor | refactoring.md — only when UI is in scope |
| Kotlin, Swift, Android/iOS mobile | language: mobile | android.md |
Precedence: missing tests on critical paths → tdd first; then explicit user mode; then detected signals.
Golden rules (short)
From the prior catalog refactor skill (legacy-principles.md):
- Behavior preserved — change how, not what (unless scoped).
- Small steps; verify after each boundary.
- One purpose per change set — no mixed features.
- Tests or agreed proof bracket structural edits.
- Skip refactor when code is stable, untested-critical, or deadline-blocked without a plan.
Default pipelines
Large change: plan → user confirms → safe-execute (repeat verify) → optional review-and-align.
Single-file cleanup: safe-execute; add reduce-complexity when one function dominates.
Go at scale: plan (inventory) → execute one atomic change per step per golang track.
Elixir: elixir track → load super-elixir; do not duplicate BEAM guides here.
Legacy catalog depth
Retained content from refactor (examples and checklists):
| Topic | Reference |
| Principles and when not to | legacy-principles.md |
| Ten smells with diffs | legacy-code-smells.md |
| Extract method, typing | legacy-extract-and-types.md |
| Strategy, chain patterns | legacy-design-patterns.md |
| Steps, checklist, operations table | legacy-steps-checklist-operations.md |
Layout
| Path | Holds |
| references/core/ | Mode routing, protocols, patterns overview, duplication, output-savers bridge, research |
| references/legacy/ | Prior catalog refactor body |
| references/tech/ | Per-stack tracks (Go, Elixir pointer, Terraform, React, web UI, mobile stub) |
Old skill name
| Old | Use |
refactor | This skill (safe-refactor) |
Related skills
tdd: add a safety net before you move code.verify: prove behavior is unchanged after each step.output-savers: keep the diff minimal.codebase-learner: map the smells before you plan.super-architecture: name the target style.code-review: review the refactor diff.atomic-semantic-commit: commit each safe step on its own.
Done when
- Correct mode(s) ran; only needed references were loaded.
- Behavior preserved per boundary definition; proof run before and after execute modes with failures explicitly checked.
- Diffs stay minimal per output-savers; no scope creep into features.
- Plan mode produced a plan and waited for confirmation before code (unless user waived).
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.