uboy/AICodingAgentsAndSkills
Shared policies, skills and runner configs for AI coding agents (Claude Code, Codex, OpenCode, ZCode)
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 uboy/AICodingAgentsAndSkills/agents ./rig-aicodingagentsandskills/agents $ npx degit uboy/AICodingAgentsAndSkills/commands ./rig-aicodingagentsandskills/commands $ npx degit uboy/AICodingAgentsAndSkills/skills ./rig-aicodingagentsandskills/skills
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 uboy/AICodingAgentsAndSkills/skills/agent-system-coach .claude/skills/agent-system-coach $ npx degit uboy/AICodingAgentsAndSkills/skills/android-code-review .claude/skills/android-code-review $ npx degit uboy/AICodingAgentsAndSkills/skills/api-contract-review .claude/skills/api-contract-review $ npx degit uboy/AICodingAgentsAndSkills/skills/build-system-analysis .claude/skills/build-system-analysis $ npx degit uboy/AICodingAgentsAndSkills/skills/case-analyzer .claude/skills/case-analyzer $ npx degit uboy/AICodingAgentsAndSkills/skills/ci-failure-debugger .claude/skills/ci-failure-debugger $ npx degit uboy/AICodingAgentsAndSkills/skills/code-review .claude/skills/code-review $ npx degit uboy/AICodingAgentsAndSkills/skills/gitcode-pr-issue .claude/skills/gitcode-pr-issue $ npx degit uboy/AICodingAgentsAndSkills/skills/gitcode-pr-review .claude/skills/gitcode-pr-review $ npx degit uboy/AICodingAgentsAndSkills/skills/homework-management .claude/skills/homework-management $ npx degit uboy/AICodingAgentsAndSkills/skills/ios-code-review .claude/skills/ios-code-review $ npx degit uboy/AICodingAgentsAndSkills/skills/java-code-review .claude/skills/java-code-review $ npx degit uboy/AICodingAgentsAndSkills/skills/large-codebase-context .claude/skills/large-codebase-context $ npx degit uboy/AICodingAgentsAndSkills/skills/lecture-transcript .claude/skills/lecture-transcript $ npx degit uboy/AICodingAgentsAndSkills/skills/meeting-notes .claude/skills/meeting-notes $ npx degit uboy/AICodingAgentsAndSkills/skills/openharmony-task-specifier .claude/skills/openharmony-task-specifier $ npx degit uboy/AICodingAgentsAndSkills/skills/performance-review .claude/skills/performance-review $ npx degit uboy/AICodingAgentsAndSkills/skills/security-review .claude/skills/security-review $ npx degit uboy/AICodingAgentsAndSkills/skills/task-progress .claude/skills/task-progress $ npx degit uboy/AICodingAgentsAndSkills/skills/task-specifier .claude/skills/task-specifier $ npx degit uboy/AICodingAgentsAndSkills/skills/text-cleanup .claude/skills/text-cleanup $ npx degit uboy/AICodingAgentsAndSkills/skills/text-humanize .claude/skills/text-humanize
$ curl -fsSL --create-dirs -o .claude/agents/agent-architect.md https://raw.githubusercontent.<redacted>-architect.md $ curl -fsSL --create-dirs -o .claude/agents/agent-lawyer.md https://raw.githubusercontent.<redacted>-lawyer.md $ curl -fsSL --create-dirs -o .claude/agents/agent-system-coach.md https://raw.githubusercontent.<redacted>-system-coach.md $ curl -fsSL --create-dirs -o .claude/agents/code-review-qa.md https://raw.githubusercontent.<redacted>-review-qa.md $ curl -fsSL --create-dirs -o .claude/agents/debug-detective.md https://raw.githubusercontent.<redacted>-detective.md $ curl -fsSL --create-dirs -o .claude/agents/devops-engineer.md https://raw.githubusercontent.<redacted>-engineer.md $ curl -fsSL --create-dirs -o .claude/agents/docs-writer.md https://raw.githubusercontent.<redacted>-writer.md $ curl -fsSL --create-dirs -o .claude/agents/homework-indexer.md https://raw.githubusercontent.<redacted>-indexer.md $ curl -fsSL --create-dirs -o .claude/agents/homework-manager.md https://raw.githubusercontent.<redacted>-manager.md $ curl -fsSL --create-dirs -o .claude/agents/implementation-developer.md https://raw.githubusercontent.<redacted>-developer.md $ curl -fsSL --create-dirs -o .claude/agents/lang-arkts.md https://raw.githubusercontent.<redacted>-arkts.md $ curl -fsSL --create-dirs -o .claude/agents/lang-cpp.md https://raw.githubusercontent.<redacted>-cpp.md $ curl -fsSL --create-dirs -o .claude/agents/lang-python.md https://raw.githubusercontent.<redacted>-python.md $ curl -fsSL --create-dirs -o .claude/agents/lang-typescript.md https://raw.githubusercontent.<redacted>-typescript.md $ curl -fsSL --create-dirs -o .claude/agents/lead-dev-planner.md https://raw.githubusercontent.<redacted>-dev-planner.md $ curl -fsSL --create-dirs -o .claude/agents/text-editor.md https://raw.githubusercontent.<redacted>-editor.md $ curl -fsSL --create-dirs -o .claude/agents/wm-implementer.md https://raw.githubusercontent.<redacted>-model/wm-implementer.md $ curl -fsSL --create-dirs -o .claude/agents/wm-orchestrator.md https://raw.githubusercontent.<redacted>-model/wm-orchestrator.md $ curl -fsSL --create-dirs -o .claude/agents/wm-planner.md https://raw.githubusercontent.<redacted>-model/wm-planner.md $ curl -fsSL --create-dirs -o .claude/agents/wm-reviewer.md https://raw.githubusercontent.<redacted>-model/wm-reviewer.md
This rig commits no guardrails. Here is the community baseline instead — the deny/ask rules most often found across all 6,974 rigs:
{
"permissions": {
"deny": [
"Read(./.env)",
"Read(**/.env)",
"Read(~/.ssh/**)",
"Bash(rm -rf *)",
"Read(**/*.pem)",
"Bash(rm -rf /)",
"Bash(git push --force:*)",
"Bash(sudo *)",
"Read(~/.aws/**)",
"Bash(rm -rf /*)",
"Read(./.env.*)",
"Read(.env)",
"Bash(git push --force*)",
"Bash(rm -rf:*)",
"Read(**/.env.*)",
"Read(**/*.key)",
"Bash(sudo:*)",
"Bash(git reset --hard*)",
"Bash(git reset --hard:*)",
"Read(.env.*)"
],
"ask": [
"Bash(git push:*)",
"Bash(git push *)",
"Bash(git commit:*)",
"Bash(rm *)",
"Bash(rm:*)",
"Bash(npm publish:*)",
"Bash(wget *)",
"Bash(git rebase *)",
"Bash(gh pr merge *)",
"Bash(git commit *)"
]
}
} Skills (22)
agent-system-coachandroid-code-reviewapi-contract-reviewbuild-system-analysiscase-analyzerci-failure-debuggercode-reviewgitcode-pr-issuegitcode-pr-reviewhomework-managementios-code-reviewjava-code-reviewlarge-codebase-contextlecture-transcriptmeeting-notesopenharmony-task-specifierperformance-reviewsecurity-reviewtask-progresstask-specifiertext-cleanuptext-humanize
Subagents (20)
| agent-architect model: opus | Use this agent when the user needs architectural guidance, feature design, codebase understanding, implementation planning, or strategic technical decisions. This includes designing new features, unde |
| agent-lawyer model: sonnet | Use this agent only when the user explicitly asks for legal/license/compliance review (license compatibility, dependency licensing, patent signals, export/compliance concerns). Do not invoke proactive |
| agent-system-coach model: sonnet | Use this agent when the user needs onboarding, coaching, or operating guidance for the repository's multi-agent workflow. It teaches the execution sequence, required verification/review commands, and |
| code-review-qa model: sonnet | Use this agent when code changes have been made and need to be reviewed for compliance with requirements, code quality, bug detection, and performance issues. This agent should be launched after any m |
| debug-detective model: opus | Use this agent when the user encounters an unknown bug, crash, stack trace, flaky test, memory leak, performance regression, or any issue that requires root cause analysis and diagnosis. This agent in |
| devops-engineer model: sonnet | Use this agent when the user needs help with CI/CD pipelines, Docker configuration, deployment setup, build systems, environment configuration, GitHub Actions, pre-commit hooks, linting setup, or deve |
| docs-writer model: sonnet | Use this agent when the user needs documentation written, updated, or improved. This includes README files, API documentation, Architecture Decision Records (ADRs), changelogs, onboarding guides, inli |
| homework-indexer model: haiku | Use this agent for indexing and searching university course materials. It scans a read-only directory of lectures, transcripts, presentations, and books, builds a structured index, and extracts archiv |
| homework-manager model: sonnet | Use this agent for end-to-end academic homework orchestration: extract assignment requirements, map sources, build a step-by-step execution plan, draft the answer section by section, and finish with a |
| implementation-developer model: sonnet | Use this agent when the user has a well-defined task, feature, or piece of work that needs to be implemented in code. This includes writing new features, refactoring existing code, fixing bugs, or bui |
| lang-arkts model: sonnet | Use this agent for ArkTS-specific advisory — ArkUI declarative UI patterns, HarmonyOS development, ArkTS limitations vs TypeScript, component lifecycle, state management decorators, and common pitfall |
| lang-cpp model: sonnet | Use this agent for C/C++-specific advisory — modern C++ patterns (C++17/20/23), memory safety, RAII, smart pointers, CMake, testing, and common pitfalls. This is an advisory agent: it provides recomme |
| lang-python model: sonnet | Use this agent for Python-specific advisory — idiomatic patterns, type hints, async best practices, testing strategies, packaging, and common pitfalls. This is an advisory agent: it provides recommend |
| lang-typescript model: sonnet | Use this agent for TypeScript/JavaScript-specific advisory — type system patterns, ESM modules, framework best practices, testing strategies, and common pitfalls. This is an advisory agent: it provide |
| lead-dev-planner model: sonnet | Use this agent when the user needs to decompose a feature into development tasks, create work breakdown structures, plan task assignments across a team, or organize development work. This includes fea |
| text-editor model: sonnet | Use this agent for high-quality text editing and transcript transformation tasks: cleanup, restructuring, lecture-note generation, and meeting summary extraction. This agent is for text workflows wher |
| wm-implementer model: haiku | Weak-model atomic coder. Use when a specific, well-defined code change is needed and the active model is constrained (claude-haiku, gpt-oss-120b, qwen3.5:27b, or similar). One file, max 30 lines of ch |
| wm-orchestrator model: haiku | Weak-model team-lead orchestrator. Use when the active model is a constrained/fast model (claude-haiku, gpt-oss-120b, qwen3.5:27b, or similar). Replaces the standard multi-agent orchestration with a s |
| wm-planner model: haiku | Weak-model lightweight task planner. Produces a simple JSON checklist of up to 7 micro-tasks. Each task targets one file with at most 30 lines of change and an explicit, testable acceptance criterion. |
| wm-reviewer model: haiku | Weak-model code reviewer. Performs exactly three checks – correct, secure, spec-compliant – on one file per invocation. JSON output only. Use when code review is needed on a constrained model.\n\nExam |
Slash commands (13)
/agent-system-coach/android-code-review/api-contract-review/ci-failure-debugger/code-review/ios-code-review/java-code-review/lecture-transcript/meeting-notes/performance-review/security-review/task-progress/text-cleanup
Similar rigs
ker0olos/ignite
The only AI coding app you'll need. Run agents in parallel. Hand off tasks that come back as pull requests. Draw on screenshots to show exactly what's wrong. Native macOS, sandboxed by default.
Pragmatist 16.8k tok ·
majidraza1228/dotfiles
Personal dotfiles — Claude Code config and install script
Automator 1.3k tok ·
tdiprima/claude-config
Claude Code dotfiles and bootstrap script to configure a new machine with rules, skills, and settings in one command.
Pragmatist 686 tok ·
selmakcby/claude-agents-skills
Multi-Agent Claude Code setup — 4 uzman ajan (planner · ui-agent · builder · reviewer) + skills + Next.js demo projesi. YouTube Bölüm 1 video materyalleri.
Orchestrator 1.5k tok ·