Skip to content
Back to search

claude-code-agents

SkillWorks with Claude CodeGitHub

Claude Code Agents Prompt templates for Claude Code's subagent system. Run parallel code audits, automate fix cycles, get stuff reviewed. Built for the Claude Code Task tool. Not a framework. Just prompts that work. What This Does Claude Code can spawn subagents via Task(). These are the prompts those agents receive.

Safety scan

A static text scan, not a guarantee. Select the badge to see what was scanned and every finding.

Checked today
Badge for your README
findsafeskills safety rating

Markdown

[![findsafeskills safety rating](/api/listings/undeadlist--claude-code-agents/badge.svg)](/listings/undeadlist--claude-code-agents)

HTML

<a href="/listings/undeadlist--claude-code-agents"><img src="/api/listings/undeadlist--claude-code-agents/badge.svg" alt="findsafeskills safety rating" /></a>

Install

i

Install instructions from the README

As written by the repo's author. findsafeskills has not verified these steps.

These steps run a script from the network that findsafeskills does not scan.

Install as Plugin (Recommended)

Install once, use across all projects. No per-project file copying needed.

# Step 1: Add the marketplace (inside Claude Code)
/plugin marketplace add undeadlist/claude-code-agents

# Step 2: Install the plugin (inside Claude Code)
/plugin install claude-code-agents@undeadlist-claude-code-agents

Or after Step 1, use `/plugin` → **Discover** tab to browse and install.

You can also install from the terminal:

claude plugin install claude-code-agents@undeadlist-claude-code-agents --scope user

All 24 agents and 6 workflow skills become natively available. Use skills like `/full-audit`, `/pre-commit`, `/pre-deploy`, etc.

Verify Your Install

After installing, confirm everything loaded:

# Inside Claude Code, ask:
"List all available agents and skills from the claude-code-agents plugin"

You should see 24 agents and 6 skills. If not, check the [Troubleshooting](#troubleshooting) section below.

Try Before Installing

Test the plugin locally without installing:

git clone https://github.com/undeadlist/claude-code-agents.git
claude --plugin-dir ./claude-code-agents

This loads the plugin for that session only — nothing is installed globally.

Alternative Installation (Per-Project)

If you prefer per-project installation:

# Option A: Full setup (creates CLAUDE.md, detects stack)
curl -s https://undeadlist.com/agents/setup-project.sh | bash

# Option B: Just agents
curl -s https://undeadlist.com/agents/install.sh | bash

# Option C: Via npm
npx claude-code-agents

This copies agents into your project's `.claude/agents/` directory.

---

What was scanned

The files the safety scan read, as written by the repository's author and shown as plain text. Lines with a finding are marked.

.claude-plugin/marketplace.json · Manifest · 24 lines

View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)

  1. {
  2. "name": "undeadlist-claude-code-agents",
  3. "owner": {
  4. "name": "UndeadList",
  5. "email": "hello@undeadlist.com"
  6. },
  7. "metadata": {
  8. "description": "24 parallel audit agents + 6 workflow skills for Claude Code",
  9. "version": "1.0.0"
  10. },
  11. "plugins": [
  12. {
  13. "name": "claude-code-agents",
  14. "source": "./",
  15. "description": "24 audit agents + 6 workflow skills for solo dev startups building with Next.js + Claude Code",
  16. "version": "3.1.0",
  17. "author": {
  18. "name": "UndeadList"
  19. },
  20. "category": "development-tools",
  21. "tags": ["audit", "security", "code-quality", "testing", "deployment", "qa", "solo-dev"]
  22. }
  23. ]
  24. }
README.md · README · 570 lines · 6 findings marked

View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)

  1. # Complete E2E Development Workflow
  2. [![Claude Code Ready](badges/claude-code-ready.svg)](https://undeadlist.com) [![Mentioned in Awesome Claude Code](https://awesome.re/mentioned-badge.svg)](https://github.com/hesreallyhim/awesome-claude-code)
  3. ## For Solo Dev Startups Building with Next.js + Claude Code
  4. ![Parallel Audit Demo](./assets/terminal-demo.svg)
  5. > **Target User:** Solo dev running a startup, only engineer, entire company.
  6. > **Goal:** Pull this repo into any project and have a full agent team ready to go.
  7. > **Designed for:** Next.js / React / TypeScript full-stack web apps (Prisma, npm/pnpm, Vercel).
  8. > Other stacks can adapt the agent prompts, but examples and tooling defaults target this stack.
  9. > **Built by:** Paul @ UndeadList — learned the hard way what works.
  10. ---
  11. ## The Reality Check
  12. You're a solo dev. You don't have:
  13. - A QA team
  14. - A code reviewer
  15. - A DevOps engineer
  16. - Time to waste on AI agents going rogue
  17. This workflow package gives you all of that via Claude Code subagents, with strict protocols to prevent the bullshit that wastes your time.
  18. ---
  19. ## What's In This Repo
  20. ```
  21. claude-code-agents/
  22. ├── .claude-plugin/
  23. │ ├── plugin.json # Plugin manifest
  24. │ └── marketplace.json # Marketplace config
  25. │
  26. ├── agents/ # 24 agent definitions
  27. │ ├── # AUDIT AGENTS (11 - run in parallel)
  28. │ ├── code-auditor.md # Code quality, DRY, complexity
  29. │ ├── bug-auditor.md # Runtime bugs, auth gaps
  30. │ ├── security-auditor.md # OWASP deep scan
  31. │ ├── doc-auditor.md # Documentation gaps
  32. │ ├── infra-auditor.md # Config, env vars, headers
  33. │ ├── ui-auditor.md # Accessibility, UX
  34. │ ├── db-auditor.md # Database, N+1, indexes
  35. │ ├── perf-auditor.md # Performance, bundle size
  36. │ ├── dep-auditor.md # Dependencies, vulnerabilities
  37. │ ├── seo-auditor.md # SEO, meta tags, OpenGraph
  38. │ ├── api-tester.md # API endpoint testing
  39. │ ├── # FIX/IMPLEMENT AGENTS (4)
  40. │ ├── fix-planner.md # Prioritizes findings into FIXES.md
  41. │ ├── code-fixer.md # Implements fixes
  42. │ ├── test-runner.md # Validates fixes
  43. │ ├── test-writer.md # Auto-generates tests
  44. │ ├── # BROWSER AGENTS (4 - Chrome integration)
  45. │ ├── browser-qa-agent.md # Navigates UI, finds console errors
  46. │ ├── fullstack-qa-orchestrator.md # Find → Fix → Verify loop
  47. │ ├── console-monitor.md # Real-time console watching
  48. │ ├── visual-diff.md # Screenshot comparison
  49. │ ├── # DEPLOY AGENTS (2)
  50. │ ├── deploy-checker.md # Pre-deployment validation
  51. │ ├── env-validator.md # Environment configuration
  52. │ ├── # UTILITY AGENTS (2)
  53. │ ├── pr-writer.md # PR description generator
  54. │ ├── seed-generator.md # Test data creation
  55. │ └── # SUPERVISORS (1)
  56. │ └── architect-reviewer.md # Oversees until production-ready
  57. │
  58. ├── skills/ # 6 workflow skills (slash commands)
  59. │ ├── full-audit/SKILL.md # /full-audit
  60. │ ├── pre-commit/SKILL.md # /pre-commit
  61. │ ├── pre-deploy/SKILL.md # /pre-deploy
  62. │ ├── new-feature/SKILL.md # /new-feature
  63. │ ├── bug-fix/SKILL.md # /bug-fix
  64. │ └── release-prep/SKILL.md # /release-prep
  65. │
  66. ├── workflows/ # Detailed workflow documentation
  67. │ ├── pre-commit.md
  68. │ ├── pre-deploy.md
  69. │ ├── full-audit.md
  70. │ ├── new-feature.md
  71. │ ├── bug-fix.md
  72. │ └── release-prep.md
  73. │
  74. ├── CLAUDE.md # Orchestrator instructions
  75. ├── setup-project.sh # One command setup for new projects
  76. ├── install.sh # Quick agent install (legacy)
  77. └── README.md
  78. ```
  79. **Total: 24 agents + 6 workflow skills**
  80. ---
  81. ## Quick Start
  82. ### Install as Plugin (Recommended)
  83. Install once, use across all projects. No per-project file copying needed.
  84. ```bash
  85. # Step 1: Add the marketplace (inside Claude Code)
  86. /plugin marketplace add undeadlist/claude-code-agents
  87. # Step 2: Install the plugin (inside Claude Code)
  88. /plugin install claude-code-agents@undeadlist-claude-code-agents
  89. ```
  90. Or after Step 1, use `/plugin` → **Discover** tab to browse and install.
  91. You can also install from the terminal:
  92. ```bash
  93. claude plugin install claude-code-agents@undeadlist-claude-code-agents --scope user
  94. ```
  95. All 24 agents and 6 workflow skills become natively available. Use skills like `/full-audit`, `/pre-commit`, `/pre-deploy`, etc.
  96. ### Verify Your Install
  97. After installing, confirm everything loaded:
  98. ```bash
  99. # Inside Claude Code, ask:
  100. "List all available agents and skills from the claude-code-agents plugin"
  101. ```
  102. You should see 24 agents and 6 skills. If not, check the [Troubleshooting](#troubleshooting) section below.
  103. ### Try Before Installing
  104. Test the plugin locally without installing:
  105. ```bash
  106. git clone https://github.com/undeadlist/claude-code-agents.git
  107. claude --plugin-dir ./claude-code-agents
  108. ```
  109. This loads the plugin for that session only — nothing is installed globally.
  110. ### Alternative Installation (Per-Project)
  111. If you prefer per-project installation:
  112. ```bash
  113. # Option A: Full setup (creates CLAUDE.md, detects stack)
  114. curl -s https://undeadlist.com/agents/setup-project.sh | bash
    • Note: Pipes a downloaded script directly into a shell (curl | sh / wget | bash)
    • Line 146 on GitHub
  115. # Option B: Just agents
  116. curl -s https://undeadlist.com/agents/install.sh | bash
    • Note: Pipes a downloaded script directly into a shell (curl | sh / wget | bash)
    • Line 149 on GitHub
  117. # Option C: Via npm
  118. npx claude-code-agents
  119. ```
  120. This copies agents into your project's `.claude/agents/` directory.
  121. ---
  122. ## Stack Compatibility
  123. These agents are optimized for **Next.js / React / TypeScript** projects using **Prisma**, **npm/pnpm**, and **Vercel**. Commands, file paths, and examples throughout the agent prompts reflect this stack.
  124. **Stack-agnostic agents** (useful for any project):
  125. - `security-auditor` — OWASP checks apply to any web app
  126. - `code-auditor` — quality principles are language-agnostic
  127. - `doc-auditor` — documentation gap analysis works everywhere
  128. - `pr-writer` — generates PR descriptions for any git repo
  129. - `fix-planner` — prioritization logic is framework-independent
  130. - `architect-reviewer` — review criteria are generic
  131. **Adapting for other stacks:** Fork the repo and edit the agent `.md` files in `agents/`. Swap Next.js-specific commands (`npm run dev`, `pnpm tsc`, `localhost:3000`) for your equivalents. The orchestration structure works regardless of stack — only the tool-specific examples need changing.
  132. ---
  133. ## Workflows
  134. ### Pre-Commit
  135. ```
  136. claude "Run pre-commit workflow"
  137. ```
  138. Runs code-auditor + test-runner before committing.
  139. ### Pre-Deploy
  140. ```
  141. claude "Run pre-deploy workflow"
  142. ```
  143. Validates build, env vars, dependencies before deployment.
  144. ### Full Audit
  145. ```
  146. claude "Run full-audit workflow on src/"
  147. ```
  148. All 11 auditors in parallel → fix-planner creates FIXES.md.
  149. ### New Feature
  150. ```
  151. claude "Run new-feature workflow for: [feature description]"
  152. ```
  153. test-writer → code-fixer → test-runner → browser-qa (TDD approach).
  154. ### Bug Fix
  155. ```
  156. claude "Run bug-fix workflow for: [bug description]"
  157. ```
  158. Write failing test → fix → verify (regression prevention).
  159. ### Release Prep
  160. ```
  161. claude "Run release-prep workflow for v1.0.0"
  162. ```
  163. Full audit → fixes → deploy validation → PR generation.
  164. ---
  165. ## Project Lifecycle
  166. ### Phase 0: New Project Setup
  167. ```bash
  168. curl -s https://undeadlist.com/agents/setup-project.sh | bash
    • Note: Pipes a downloaded script directly into a shell (curl | sh / wget | bash)
    • Line 220 on GitHub
  169. ```
  170. ### Phase 1: Skeleton/Scaffolding
  171. **Before touching code, create your blueprint:**
  172. ```
  173. claude "Create a SETUP_BLUEPRINT.md for this project:
  174. - Folder structure
  175. - Tech stack decisions (with rationale)
  176. - Dependencies list (don't install yet)
  177. - Environment variables needed
  178. - Database planning
  179. - Success criteria checklist
  180. Do NOT generate any code. Just the plan."
  181. ```
  182. ### Phase 2: Initial Build
  183. **Strict rules for AI agents during build:**
  184. ```markdown
  185. ## RULES FOR THIS SESSION
  186. 1. **ASK BEFORE** modifying any file not directly related to current task
  187. 2. **NEVER** refactor "while you're in there"
  188. 3. **NEVER** update dependencies unless explicitly requested
  189. 4. **NEVER** change config files without approval
  190. 5. **ONE logical change per commit**
  191. 6. **VERIFY** the build works after each change
  192. ```
  193. ### Phase 3: Development Loop
  194. The core loop for any task:
  195. ```
  196. ┌─────────────────────────────────────────────────┐
  197. │ 1. Understand the task │
  198. │ 2. State what files you'll touch │
  199. │ 3. Get approval │
  200. │ 4. Make ONE change │
  201. │ 5. Verify it works │
  202. │ 6. Commit │
  203. │ 7. Repeat │
  204. └─────────────────────────────────────────────────┘
  205. ```
  206. ### Phase 4: Pre-Launch Audit
  207. Run the full parallel audit:
  208. ```
  209. claude "Run full-audit workflow on src/"
  210. ```
  211. **Review `.claude/audits/` for:**
  212. - AUDIT_CODE.md, AUDIT_SECURITY.md, AUDIT_DB.md, AUDIT_PERF.md
  213. - FIXES.md (prioritized action items)
  214. ### Phase 5: Browser QA
  215. Enable Chrome integration:
  216. ```
  217. /chrome
  218. ```
  219. Run the full QA loop:
  220. ```
  221. claude "Use fullstack-qa-orchestrator to:
  222. 1. Start dev server (npm run dev)
  223. 2. Navigate to http://localhost:3000
  224. 3. Test: homepage, auth flow, main features
  225. 4. Find console errors, UI bugs
  226. 5. Fix issues found
  227. 6. Re-test to verify
  228. 7. Repeat until clean"
  229. ```
  230. ### Phase 6: Deploy
  231. ```
  232. claude "Run pre-deploy workflow"
  233. ```
  234. ---
  235. ## The Protocols
  236. ### Protocol 1: No Unauthorized Changes
  237. Every AI agent must follow:
  238. ```markdown
  239. ## BEFORE ANY CHANGE
  240. 1. State what file you're modifying
  241. 2. State what change you're making
  242. 3. State why
  243. 4. Wait for "PROCEED" or "STOP"
  244. ## FORBIDDEN ACTIONS
  245. - ❌ npm install / npx anything without approval
  246. - ❌ Modifying package.json without approval
  247. - ❌ Creating new files when editing existing works
  248. - ❌ Refactoring unrelated code
  249. - ❌ "Improvements" not requested
  250. - ❌ Touching config files without approval
  251. ```
  252. ### Protocol 2: Micro-Checkpoint Method
  253. For complex changes:
  254. ```markdown
  255. ## EVERY PHASE:
  256. 1. Agent states planned change
  257. 2. Agent calls architect-reviewer for verification
  258. 3. User says "PROCEED" or "STOP"
  259. 4. Agent makes ONE change
  260. 5. Agent verifies change applied correctly
  261. 6. Proceed to next phase only with approval
  262. ```
  263. ### Protocol 3: Regression Prevention
  264. Before any change:
  265. ```bash
  266. # Baseline testing
  267. curl -s http://localhost:3000/api/health # API works
  268. pnpm tsc --noEmit # TypeScript valid
  269. pnpm test # Tests pass
  270. # After change - all must still work
  271. ```
  272. ---
  273. ## Common Commands
  274. ### Daily Development
  275. ```bash
  276. # Start coding session
  277. claude "Review CLAUDE.md, then help me with [task]"
  278. # Quick check before commit
  279. claude "Run pre-commit workflow"
  280. # Verify a specific file
  281. claude "Review [file] for bugs, especially [concern area]"
  282. ```
  283. ### Pre-Launch
  284. ```bash
  285. # Full audit
  286. claude "Run full-audit workflow on src/"
  287. # Browser QA
  288. claude "Use fullstack-qa-orchestrator to test http://localhost:3000"
  289. # Security focus
  290. claude "Use security-auditor to deep scan src/api/"
  291. ```
  292. ### Debugging
  293. ```bash
  294. # Console monitoring
  295. claude "Use console-monitor to watch http://localhost:3000"
  296. # Specific bug
  297. claude "Run bug-fix workflow for: [bug description]"
  298. ```
  299. ---
  300. ## Files Generated
  301. All reports go to `.claude/audits/` (gitignored):
  302. | File | Source Agent |
  303. |------|--------------|
  304. | AUDIT_CODE.md | code-auditor |
  305. | AUDIT_SECURITY.md | bug-auditor |
  306. | AUDIT_SECURITY_DEEP.md | security-auditor |
  307. | AUDIT_DOCS.md | doc-auditor |
  308. | AUDIT_INFRA.md | infra-auditor |
  309. | AUDIT_UI_UX.md | ui-auditor |
  310. | AUDIT_DB.md | db-auditor |
  311. | AUDIT_PERF.md | perf-auditor |
  312. | AUDIT_DEPS.md | dep-auditor |
  313. | AUDIT_SEO.md | seo-auditor |
  314. | API_TEST_REPORT.md | api-tester |
  315. | DEPLOY_CHECK.md | deploy-checker |
  316. | ENV_REPORT.md | env-validator |
  317. | FIXES.md | fix-planner |
  318. | TEST_REPORT.md | test-runner |
  319. | AUDIT_BROWSER_QA.md | browser-qa-agent |
  320. | QA_SESSION_LOG.md | fullstack-qa-orchestrator |
  321. ---
  322. ## Quick Reference
  323. | Task | Command |
  324. |------|---------|
  325. | New project setup | `curl -s https://undeadlist.com/agents/setup-project.sh \| bash` |
    • Note: Pipes a downloaded script directly into a shell (curl | sh / wget | bash)
    • Line 433 on GitHub
  326. | Enable browser QA | `/chrome` in Claude Code |
  327. | Pre-commit check | `"Run pre-commit workflow"` |
  328. | Full audit | `"Run full-audit workflow on src/"` |
  329. | Pre-deploy check | `"Run pre-deploy workflow"` |
  330. | Browser test | `"Use browser-qa-agent to test localhost:3000"` |
  331. | Full QA loop | `"Use fullstack-qa-orchestrator..."` |
  332. | New feature | `"Run new-feature workflow for: [feature]"` |
  333. | Bug fix | `"Run bug-fix workflow for: [bug]"` |
  334. | Release prep | `"Run release-prep workflow for v1.0.0"` |
  335. | Verify production-ready | `"Use architect-reviewer..."` |
  336. ---
  337. ## When AI Goes Rogue
  338. ### Red Flags - STOP Immediately
  339. - Agent makes changes without asking
  340. - Agent installs packages without approval
  341. - Agent modifies config files unexpectedly
  342. - Agent claims success without verification
  343. - Agent skips checkpoints
  344. - Agent touches files outside scope
  345. ### Recovery Commands
  346. ```bash
  347. # Revert last change
  348. git checkout -- [file]
  349. # Revert all changes
  350. git reset --hard HEAD
  351. # Check what changed
  352. git diff
  353. ```
  354. ### Enforcement Phrases
  355. ```
  356. "STOP. You didn't ask permission."
  357. "STOP. That's not in scope."
  358. "STOP. Verify it works before claiming done."
  359. "STOP. One change at a time."
  360. ```
  361. ---
  362. ## Troubleshooting
  363. ### Plugin not loading / no agents found
  364. 1. **Check plugin errors:** Run `/plugin` inside Claude Code and check the **Errors** tab
  365. 2. **Verify marketplace was added:** Go to `/plugin` → **Marketplaces** tab — you should see `undeadlist-claude-code-agents`
  366. 3. **Verify plugin is installed:** Go to `/plugin` → **Installed** tab — you should see `claude-code-agents`
  367. 4. **Re-add and reinstall:**
  368. ```bash
  369. /plugin marketplace add undeadlist/claude-code-agents
  370. /plugin install claude-code-agents@undeadlist-claude-code-agents
  371. ```
  372. ### Skills (slash commands) not appearing
  373. Skills like `/full-audit` only work when installed as a plugin. Per-project installs (`npx claude-code-agents`) install agents to `.claude/agents/` but don't register slash commands — use natural language instead:
  374. ```
  375. claude "Run full-audit workflow on src/"
  376. ```
  377. ### `npx claude-code-agents` fails
  378. - Requires Node.js >= 14
  379. - Must be run from a project directory
  380. - Copies agents to `.claude/agents/` and workflows to `workflows/` in your project
  381. ### `curl` installer returns 404
  382. The setup script URL is `https://undeadlist.com/agents/setup-project.sh` (not `setup.sh`). If the CDN isn't configured, use the GitHub raw URL:
  383. ```bash
  384. curl -sL https://raw.githubusercontent.com/undeadlist/claude-code-agents/main/install.sh | bash
    • Note: Pipes a downloaded script directly into a shell (curl | sh / wget | bash)
    • Line 512 on GitHub
  385. ```
  386. ### Agents exist but aren't being used
  387. Make sure your project has a `CLAUDE.md` that references the agents. Run `setup-project.sh` to auto-generate one, or see the [CLAUDE.md](CLAUDE.md) in this repo as a template.
  388. ---
  389. ## Uninstall
  390. ### Plugin
  391. ```bash
  392. # In Claude Code:
  393. /plugin uninstall claude-code-agents
  394. ```
  395. ### Per-Project Install
  396. ```bash
  397. # Option A: One-liner (recommended)
  398. curl -s https://raw.githubusercontent.com/undeadlist/claude-code-agents/main/uninstall.sh | bash
    • Note: Pipes a downloaded script directly into a shell (curl | sh / wget | bash)
    • Line 532 on GitHub
  399. # Option B: Manual removal
  400. rm -rf .claude/agents/ .claude/audits/ workflows/
  401. ```
  402. This removes:
  403. - `.claude/agents/` - agent prompt definitions
  404. - `.claude/audits/` - generated reports (gitignored)
  405. - `workflows/` - predefined workflow chains
  406. > **Note:** CLAUDE.md is left in place. Remove manually if desired.
  407. ---
  408. ## Contributing
  409. Found a better protocol? Agent that should exist?
  410. PR it: https://github.com/undeadlist/claude-code-agents
  411. ---
  412. ## License
  413. MIT. Use it, fork it, enhance it.
  414. ---
  415. **Built by [UndeadList](https://undeadlist.com)**
  416. *The indie software flea market*
  417. ---
  418. **Disclaimer:** Claude Code is a product of [Anthropic PBC](https://anthropic.com). This project is not affiliated with or endorsed by Anthropic.
  419. ---
  420. *"Built, but undiscovered."*
install.sh · Bundled script · 111 lines · 1 finding marked
  1. #!/bin/bash
  2. # claude-code-agents installer
  3. # Usage: curl -s https://undeadlist.com/agents/install.sh | bash
    • Note: Pipes a downloaded script directly into a shell (curl | sh / wget | bash)
    • Line 4 on GitHub
  4. set -e
  5. REPO_URL="https://github.com/undeadlist/claude-code-agents"
  6. BRANCH="main"
  7. echo "claude-code-agents installer"
  8. echo "================================"
  9. echo ""
  10. # Check if we're in a project directory
  11. if [ ! -f "package.json" ] && [ ! -f "Cargo.toml" ] && [ ! -f "go.mod" ] && [ ! -f "requirements.txt" ]; then
  12. echo "Warning: No project file detected (package.json, Cargo.toml, etc.)"
  13. echo " Are you in your project root?"
  14. read -p " Continue anyway? [y/N] " -n 1 -r
  15. echo
  16. if [[ ! $REPLY =~ ^[Yy]$ ]]; then
  17. exit 1
  18. fi
  19. fi
  20. # Create directories
  21. echo "Creating directories..."
  22. mkdir -p .claude/agents
  23. mkdir -p .claude/audits
  24. mkdir -p workflows
  25. # Download agent files (24 total)
  26. echo "Downloading agents..."
  27. AGENTS=(
  28. # Audit agents (11)
  29. "code-auditor.md"
  30. "bug-auditor.md"
  31. "security-auditor.md"
  32. "doc-auditor.md"
  33. "infra-auditor.md"
  34. "ui-auditor.md"
  35. "db-auditor.md"
  36. "perf-auditor.md"
  37. "dep-auditor.md"
  38. "seo-auditor.md"
  39. "api-tester.md"
  40. # Fix/Implement agents (4)
  41. "fix-planner.md"
  42. "code-fixer.md"
  43. "test-runner.md"
  44. "test-writer.md"
  45. # Browser QA agents (4)
  46. "browser-qa-agent.md"
  47. "fullstack-qa-orchestrator.md"
  48. "console-monitor.md"
  49. "visual-diff.md"
  50. # Deploy agents (2)
  51. "deploy-checker.md"
  52. "env-validator.md"
  53. # Utility agents (2)
  54. "pr-writer.md"
  55. "seed-generator.md"
  56. # Supervisors (1)
  57. "architect-reviewer.md"
  58. )
  59. for agent in "${AGENTS[@]}"; do
  60. echo " -> $agent"
  61. curl -sL "${REPO_URL}/raw/${BRANCH}/agents/${agent}" -o ".claude/agents/${agent}"
  62. done
  63. # Download workflow files (6 total)
  64. echo ""
  65. echo "Downloading workflows..."
  66. WORKFLOWS=(
  67. "pre-commit.md"
  68. "pre-deploy.md"
  69. "full-audit.md"
  70. "new-feature.md"
  71. "bug-fix.md"
  72. "release-prep.md"
  73. )
  74. for workflow in "${WORKFLOWS[@]}"; do
  75. echo " -> $workflow"
  76. curl -sL "${REPO_URL}/raw/${BRANCH}/workflows/${workflow}" -o "workflows/${workflow}"
  77. done
  78. # Create .gitignore for audits
  79. if [ ! -f ".claude/.gitignore" ]; then
  80. echo "audits/" > .claude/.gitignore
  81. fi
  82. echo ""
  83. echo "================================"
  84. echo "Installed ${#AGENTS[@]} agents to .claude/agents/"
  85. echo "Installed ${#WORKFLOWS[@]} workflows to workflows/"
  86. echo ""
  87. echo "Quick start:"
  88. echo " claude \"Run full-audit workflow on src/\""
  89. echo " claude \"Run pre-commit workflow\""
  90. echo " claude \"Run pre-deploy workflow\""
  91. echo ""
  92. echo "Tip: Install as a plugin instead (once, works across all projects):"
  93. echo " /plugin marketplace add undeadlist/claude-code-agents"
  94. echo " claude plugin install claude-code-agents@undeadlist-claude-code-agents"
  95. echo ""
  96. echo "Docs: https://github.com/undeadlist/claude-code-agents"
  97. echo "Built by UndeadList: https://undeadlist.com"
uninstall.sh · Bundled script · 48 lines · 1 finding marked
  1. #!/bin/bash
  2. # claude-code-agents uninstaller
  3. # Usage: curl -s https://raw.githubusercontent.com/undeadlist/claude-code-agents/main/uninstall.sh | bash
    • Note: Pipes a downloaded script directly into a shell (curl | sh / wget | bash)
    • Line 3 on GitHub
  4. set -e
  5. echo "claude-code-agents uninstaller"
  6. echo "================================"
  7. echo ""
  8. removed=0
  9. # Remove agent definitions
  10. if [ -d ".claude/agents" ]; then
  11. rm -rf .claude/agents/
  12. echo "Removed .claude/agents/"
  13. ((removed++))
  14. fi
  15. # Remove generated reports
  16. if [ -d ".claude/audits" ]; then
  17. rm -rf .claude/audits/
  18. echo "Removed .claude/audits/"
  19. ((removed++))
  20. fi
  21. # Remove workflows
  22. if [ -d "workflows" ]; then
  23. rm -rf workflows/
  24. echo "Removed workflows/"
  25. ((removed++))
  26. fi
  27. # Clean up empty .claude directory if nothing else is in it
  28. if [ -d ".claude" ] && [ -z "$(ls -A .claude 2>/dev/null)" ]; then
  29. rmdir .claude
  30. echo "Removed empty .claude/"
  31. fi
  32. echo ""
  33. if [ $removed -eq 0 ]; then
  34. echo "Nothing to remove. Are you in the right project directory?"
  35. else
  36. echo "================================"
  37. echo "Uninstall complete."
  38. echo ""
  39. echo "Note: CLAUDE.md was left in place (remove manually if desired)."
  40. fi

Ratings and comments

Sign in to rate this skill.

Sign in

Free to use. If it’s helped, pass it on.