Chalk Labs
Need help deploying code-cleanup?
Chalk Labs sets up agents like this inside real teams: connected to your tools, tested on your data, with a human in the loop where it matters.
Install
Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.
npx agentmarket install skill/3rdai-admin-code-cleanup
Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.
Get the weekly diff
The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.
About
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
Permissions: not checked yet.
What's inside
- Skill name
code-cleanup- Description
- None
- License
- MIT
- Instructions
- 68 lines in SKILL.md · view on GitHub
Agent Skills spec issues
- missing YAML frontmatter
- name: missing
- description: missing
- files: referenced but missing: scripts/07-validate-harness.sh
Files (2)
- SKILL.md
- code-cleanup.md
SKILL.md (first lines)
# Skill: code-cleanup
Invoked by: `/code-cleanup` command (a.k.a. **Code Cleanup**)
## Procedure
Load and follow `skills/code-cleanup/code-cleanup.md` when executing this skill.
## Purpose
Review a repository for cleanup and re-organize it to a professional, industry-standard
layout **without impacting or breaking any functionality or capabilities**. Files that are
no longer necessary are *archived* (moved to `archive/`), never deleted, so anything can be
restored if an issue surfaces. Duplicates, redundancies, and trailing artifacts are
consolidated, files are moved to appropriate folders, and the result is documented so a
developer or agent can understand the codebase.
## When to Use
- The repo has accumulated stray files, scratch scripts, duplicate configs, or ad-hoc folders
- Before a release or handoff, to present a clean, navigable structure
- After a burst of rapid iteration that left the tree disorganized
- Anytime the user asks to "clean up", "re-organize", or "tidy" the codebase
## Output
- A **cleanup plan** (proposed moves/archives/consolidations) presented for review before changes
- Files moved to appropriate folders; obsolete-but-restorable files relocated to `archive/` with a manifest
- Updated comments / READMEs so structure is self-explanatory
- A verification report confirming **no functionality changed** (references checked, tests run)
- A list of **efficiency improvements** to apply *after* cleanup (recommendations only, not applied)
## Safety Contract (Never Break Functionality)
- **Archive, don't delete.** Obsolete files move to `archive/` preserving their relative path.
- **Impact-gate every move.** Run `gitnexus_impact({target: "<symbolOrModule>", direction: "upstream"})`
before relocating any code file. Report blast radius. **HIGH or CRITICAL risk → stop and warn the user;
do not move.** For non-code files (docs, configs, assets), use `git grep -n "<filename>"` to find references.
- **Moves use the call graph.** For code symbols, prefer `gitnexus_rename` / refactoring-aware moves so
imports update; never blind find-and-replace.
- **Verify after each batch.** Run `gitnexus_detect_changes()` and the test suite; only proceed if scope
matches intent.
- **No behavior edits.** This skill moves, consolidates, and documents — it does not refactor
logic. Logic improvements are *noted*, not applied.
## Harness References
- Procedure: `skills/code-cleanup/code-cleanup.md`
- Impact analysis: `.claude/skills/gitnexus/gitnexus-impact-analysis/SKILL.md`
- Refactoring (safe moves/renames): `.claude/skills/gitnexus/gitnexus-refactoring/SKILL.md`
- Validation: `skills/validate/validate.md`, `scripts/07-validate-harness.sh`
- Commit (optional, after approval): `skills/commit/` (`/gitcommit`)
- Archive location: `archive/` (repo root)
## Autonomy Mode Recommendation
When running cleanup workflows via orchestrator:
```bash
python3 scripts/orchestrate.py route iteration --execute --adapter cli --autonomy cautious
```
**Cautious mode** is recommended for cleanup workflows because:
- Auto-approves LOW/MEDIUM risk operations (file moves, archive operations)
- Prompts for HIGH risk operations (large refactors, breaking moves)
- Blocks CRITICAL operations (force operations, mass deletions)
Review `runs/autonomy-decisions.jsonl` after completion to verify appropriate risk classifications.
## Next Step
- Success: run `/validate` (or the project test suite) to confirm health, then commit on a branch.
- If impact analysis flags HIGH/CRITICAL risk: stop, report blast radius, and hand back to the user for a decision.From 3rdAI-admin/th3rdai-harness, MIT.
More from 3rdAI-admin/th3rdai-harness
- 0.294score
security
3rdAI-admin
No description from the publisher yet.
SkillBFreshness: B★ 1Claude Code +9 - 0.294score
eval
3rdAI-admin
No description from the publisher yet.
SkillBFreshness: B★ 1Claude Code +9 - 0.264score
validate
3rdAI-admin
No description from the publisher yet.
SkillCFreshness: C★ 1Claude Code +9 - 0.264score
commit
3rdAI-admin
No description from the publisher yet.
SkillCFreshness: C★ 1Claude Code +9 - 0.264score
debug
3rdAI-admin
No description from the publisher yet.
SkillCFreshness: C★ 1Claude Code +9 - 0.264score
plan
3rdAI-admin
No description from the publisher yet.
SkillCFreshness: C★ 1Claude Code +9 - 0.264score
prompt
3rdAI-admin
No description from the publisher yet.
SkillCFreshness: C★ 1Claude Code +9 - 0.264score
research
3rdAI-admin
No description from the publisher yet.
SkillCFreshness: C★ 1Claude Code +9 - 0.264score
revise
3rdAI-admin
No description from the publisher yet.
SkillCFreshness: C★ 1Claude Code +9 - 0.264score
build
3rdAI-admin
No description from the publisher yet.
SkillCFreshness: C★ 1Claude Code +9 - 0.264score
run
3rdAI-admin
No description from the publisher yet.
SkillCFreshness: C★ 1Claude Code +9 - 0.264score
skill-script-audit
3rdAI-admin
No description from the publisher yet.
SkillCFreshness: C★ 1Claude Code +9
Signals
- Stars
- 1
- Forks
- 0
- Open issues
- 0
- Last commit (this folder)
- Jun 19, 2026 (4 months ago)
- Last release
- Jun 19, 2026 (4 months ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- MIT
- Repository
- github.com/3rdAI-admin/th3rdai-harness
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
Done for you
Rather not set it up yourself?
Chalk Labs scopes, builds and runs agents for your team: the task audit, the eval harness, the integrations and the weekly ops report. Fixed scope, priced before we start.
See how we build AI agents →Read the case studies
Chalk Labs is a two-founder AI and growth studio. We build AI agents, CRMs and MVPs for Web3 and AI companies, and have shipped 50+ product and token launches.
- launches shipped
- 50+
- launches shipped
- media placements
- 500+
- media placements
- founders on every account
- 2
- founders on every account
Press we've placed: Forbes · CoinDesk · Decrypt · Cointelegraph · The Block
Rahil Jain, Co-founder, Chalk Labs. Ex-Goldman Sachs engineer. You talk to the person who builds it.
Get code-cleanup running in your business
Tell us what you want to automate. A founder replies within one working day with a plan, not a pitch deck.
Cite this page
code-cleanup skill. Agent Mart. https://agentmart.dev/skills/3rdai-admin-code-cleanup (last updated 2026-10-24)