Coordinate multi-LLM advisory reviews of code and architecture across providers.
npx skills add https://github.com/peteski22/claude-pragma --skill star-chamber
Please help me install this Agent Skill. Skill: star-chamber Source: https://github.com/peteski22/claude-pragma/tree/main/skills/advisory/star-chamber Command: npx skills add https://github.com/peteski22/claude-pragma --skill star-chamber
The Star-Chamber skill coordinates advisory multi-LLM reviews of code and architecture by distributing prompts across providers and aggregating their feedback into a coherent recommendation.
Pipe your review prompt into the skill and run it against the configured providers. Examples include:
High-intent search queries and answers about installing and using this skill.
Multi-LLM code reviews coordinate advisory evaluations by distributing prompts across multiple providers. The system aggregates their responses to surface a consolidated set of findings and recommendations without enforcing blockers.
You need uv run to manage provider SDKs, a providers.json configuration file, and either any-llm.ai platform keys or individual provider API keys. These components are required for the skill to distribute prompts and operate correctly.
Yes, consensus code review supports both ANY_LLM_KEY platform keys and direct individual provider API keys. You configure these credentials in a providers.json file for flexible deployment across multiple providers.
Debate mode enables multi-round deliberation by allowing configured providers to evaluate prompts in parallel and structured debate formats. You can manually trigger this mode using the debate and rounds options in Claude Code to surface diverse perspectives.
No, the consensus code review skill is advisory only. It surfaces diverse perspectives and consolidated findings without enforcing blockers, allowing you to review aggregated feedback without halting your pipeline.
trueraInstrument LLM apps to capture traces for evaluation and debugging.
pradyumnasagarAutomate inter-skill data transfer and provenance capture via JSON Material Passports.
akillnessOrchestrate AI prompts across multiple providers using reusable Patterns via Unix-style piping.
GeargrindadminSimulate a five-member expert board to deliberate and reach consensus on major decisions.
