Claude Code skills for multi-agent orchestration, multi-LLM consultation, and quality auditing.
| Skill | Type | Description |
|---|---|---|
/team-play |
Command | Multi-agent orchestrated task execution with verification gates |
/deep-dive-task |
Skill | Multi-AI consultation for structured TASK document creation |
/llm-review |
Skill | Multi-LLM consultation wrapper (Gemini + GPT) |
/pikes-filter |
Skill | Rob Pike 5 principles — code complexity & over-engineering audit |
/claude-filter |
Skill | Production agent design principles — prompt/CLAUDE.md/skill audit |
/codex |
Skill | Codex CLI delegation standard — implementation handoff guide (gpt-5.3-codex) |
git clone https://github.com/TeamSPWK/nova-algorithm.git ~/.nova-algorithm
cd ~/.nova-algorithm
bash install.shRestart Claude Code after installation.
/llm-review and /deep-dive-task require external LLM API keys:
# Add to your shell profile (~/.bashrc or ~/.zshrc)
export GEMINI_API_KEY="your-gemini-api-key"
export OPENAI_API_KEY="your-openai-api-key"cd ~/.nova-algorithm
bash uninstall.shcd ~/.nova-algorithm
git pullSymlinks update automatically - no reinstall needed.