vishalguptax/agent-engineering-kit
Make AI coding agents ship clean code: rules, sub-agents, skills and hooks for Claude Code, Codex, Cursor, Copilot, Gemini CLI and 13 more, with a safe installer.
ARCHETYPE
Orchestrator
A bench of specialised subagents. The main agent mostly delegates.
Copy this rig
# review before running: this installs third-party code
$ npx degit vishalguptax/agent-engineering-kit/.claude ./rig-agent-engineering-kit # inspect, then merge into .claude/ MCP servers are added to Claude Code at local scope; env vars are shown as YOUR_… placeholders — we never store values. Files are fetched with degit into a separate folder so you can review before merging.
$ npx degit vishalguptax/agent-engineering-kit/.claude/skills/feature .claude/skills/feature $ npx degit vishalguptax/agent-engineering-kit/.claude/skills/learn .claude/skills/learn $ npx degit vishalguptax/agent-engineering-kit/.claude/skills/project-conventions .claude/skills/project-conventions $ npx degit vishalguptax/agent-engineering-kit/.claude/skills/quick-fix .claude/skills/quick-fix $ npx degit vishalguptax/agent-engineering-kit/.claude/skills/ship .claude/skills/ship $ npx degit vishalguptax/agent-engineering-kit/.claude/skills/verify-change .claude/skills/verify-change
$ curl -fsSL --create-dirs -o .claude/agents/code-architect.md https://raw.githubusercontent.com/vishalguptax/agent-engineering-kit/main/.claude/agents/code-architect.md $ curl -fsSL --create-dirs -o .claude/agents/code-explorer.md https://raw.githubusercontent.com/vishalguptax/agent-engineering-kit/main/.claude/agents/code-explorer.md $ curl -fsSL --create-dirs -o .claude/agents/code-simplifier.md https://raw.githubusercontent.com/vishalguptax/agent-engineering-kit/main/.claude/agents/code-simplifier.md $ curl -fsSL --create-dirs -o .claude/agents/security-reviewer.md https://raw.githubusercontent.com/vishalguptax/agent-engineering-kit/main/.claude/agents/security-reviewer.md $ curl -fsSL --create-dirs -o .claude/agents/silent-failure-hunter.md https://raw.githubusercontent.com/vishalguptax/agent-engineering-kit/main/.claude/agents/silent-failure-hunter.md $ curl -fsSL --create-dirs -o .claude/agents/test-analyzer.md https://raw.githubusercontent.com/vishalguptax/agent-engineering-kit/main/.claude/agents/test-analyzer.md $ curl -fsSL --create-dirs -o .claude/agents/verifier.md https://raw.githubusercontent.com/vishalguptax/agent-engineering-kit/main/.claude/agents/verifier.md
Merge into .claude/settings.json (project) or ~/.claude/settings.json (user). Hook commands reference scripts in the source repo — copy those too.
{
"permissions": {
"deny": [
"Read(./.env)",
"Read(./.env.*)",
"Read(./**/.env)",
"Read(./**/.env.*)",
"Read(./secrets/**)"
]
}
} Skills (6)
Subagents (7)
| code-architect model: inherit | Designs how to implement a feature or change so it fits the existing codebase - files to touch, interfaces, data flow, test plan, and build order in small verifiable slices. Use PROACTIVELY after expl |
| code-explorer model: inherit | Maps how an existing feature or area of the codebase works - entry points, data flow, key files, patterns to reuse - before anything is designed or changed. Use PROACTIVELY at the start of a feature o |
| code-simplifier model: inherit | Simplifies and cleans up code AFTER it works, without changing behavior. Use PROACTIVELY once an implementation passes its checks, before review. Only touches code changed in the current task. |
| security-reviewer model: inherit | Security auditor for the current change. Use PROACTIVELY for changes touching auth, user input, APIs, databases, file handling, payments, secrets/config, dependencies, or anything internet-facing. Rea |
| silent-failure-hunter model: inherit | Finds errors that are swallowed, hidden, or turned into wrong results in the current change - empty catches, ignored return codes, unawaited promises, fallbacks that mask failures. Use PROACTIVELY whe |
| test-analyzer model: inherit | Checks whether the tests for the current change actually prove it works - every changed behavior covered, edge cases and error paths tested, no weakened or flaky tests. Use PROACTIVELY before declarin |
| verifier model: inherit | Verifies that a change actually works by running the project's real checks and exercising real behavior. Use PROACTIVELY before declaring any task done. Reports evidence; does not fix code unless aske |
Hooks (1)
| event | matcher | runs |
|---|---|---|
| PostToolUse | Edit|MultiEdit|Write | node "$CLAUDE_PROJECT_DIR/.agent-kit/format.mjs" --from claude |
Permissions
deny (5)
Read(./.env)
Read(./.env.*)
Read(./**/.env)
Read(./**/.env.*)
Read(./secrets/**)
ask (0)
—
allow (0)
—
Similar rigs
tf1993614/Know-your-fate
命数天问(Mingshu Tianwen) 是一个结合 Claude Code Agent Skills、本地 Python 确定性脚本、传统数术典籍与现代 Web 对话界面 的开源 AI 命理应用。项目内置六个可离线运行的智能占卜技能,覆盖 四柱八字排盘、周公解梦、黄历择日、占事吉凶、塔罗占星、新生儿起名、风水堪舆 等场景。
Orchestrator 8.8k tok · · Q8
Iski08/dotclaude
Multi-agent config for Claude Code with specialized agents and templates to speed code review, refactoring, security audits, tech-lead guidance, and UX evaluations across CI pipelines. 🐙
Orchestrator 677 tok · · Q3
liatrio-labs/claude-code-gauntlet
Adversarial code review for Claude Code: each GitHub PR or GitLab MR runs a gauntlet of up to seven specialist agents for bugs, security, tests and cross-file impact, and every finding must survive verification, a skeptical validator, and a
Orchestrator 1.6k tok · · Q18
0xquinto/bcherny-claude
Boris Cherny's Claude Code configuration - commands, agents, and settings
Orchestrator 1.1k tok · · Q15