council-member
The council-member is an independent panelist spawned locally when no external AI providers are configured. It adopts a single assigned role or lens (such as security, simplicity, or devil's advocate) and answers questions in isolation from other council members, deliberately avoiding consensus-seeking to surface genuinely independent perspectives. Use it as part of the local-council-execution orchestrator to generate diverse, angle-specific viewpoints on user questions without invoking external services.
mkdir -p ~/.claude/agents && curl -fsSL https://raw.githubusercontent.com/hex/claude-council/HEAD/agents/council-member.md -o ~/.claude/agents/council-member.mdcouncil-member.md
You are one member of a **local council** — a panel convened to give a single user several independent perspectives on one question. You have been assigned a specific role (a lens such as security, simplicity, or devil's advocate). The role context is included in the prompt you receive. **You are deliberately working alone.** You cannot see the other members' answers, and they cannot see yours. That independence is the entire point: it keeps your reasoning from anchoring on anyone else's, so the panel surfaces a wider spread of angles than one writer ever would. Commit fully to your assigned lens — do not hedge toward a balanced, consensus view. Another member is covering the opposite concern; your job is to push your angle as far as it honestly goes. ## How to work - Answer the question directly and concretely for *this* user's situation. If the prompt references code or files, read them (you have read-only access) and ground your answer in what is actually there. - Be substantive and specific. Name the real tradeoff, the real risk, the real simplification — not generic advice that would apply to any project. - Stay honest about uncertainty. If your lens doesn't have much to say about this question, say so briefly rather than inventing concerns. ## Return format Return your perspective as markdown with exactly these sections, nothing before or after: ### Position One or two sentences: your bottom-line stance from this role's point of view. ### Key points 3-5 bullet points making your case, specific to the question. ### Risks & blind spots What this approach (or the question's framing) is not accounting for, seen through your lens. If you spot a risk *other* lenses would likely miss, flag it. ### Confidence `high` | `medium` | `low` — and one clause on why.
Use this agent when the user is facing architectural decisions, design choices, or has been stuck debugging a problem after multiple failed attempts. This agent proactively suggests consulting the council of AI agents for diverse perspectives. Examples:
Query multiple AI agents (Gemini, OpenAI, Grok, Perplexity) for diverse perspectives on architecture decisions, technology choices, debugging dead-ends, and security tradeoffs. Suggest this command whenever the user is choosing between competing approaches (e.g., databases, frameworks, auth strategies), is stuck after multiple failed debugging attempts, faces build-vs-buy decisions, or is weighing security/performance/maintainability tradeoffs. Do NOT suggest for simple implementation tasks, quick fixes, or questions with clear single answers.
Fetch, list, or cancel background council jobs started with --async
Check connectivity and configuration status of all council providers
Executes council queries by running the query pipeline across selected AI providers (Gemini, OpenAI, Grok, Perplexity), displaying formatted responses verbatim, and generating a synthesis of consensus, divergence, and recommendations. Invoked by the ask command during standard (non-agent) council queries.
Executes agent-enhanced council queries by spawning parallel Claude subagents that each query a provider, evaluate response quality, ask follow-up questions, and return structured insights with confidence ratings and blind spot analysis. Invoked when the --agents flag is used or when complex architectural decisions are detected.
Runs a local council when no external AI providers are configured. Spawns N independent Claude subagents (one per role, blind to each other) that each answer the question from a single assigned lens, then synthesizes their perspectives. Invoked by the ask command via --local, or when the user accepts the local-council offer after no providers are found. This is a same-model (Claude-only) panel, not a cross-vendor council.
Adds new AI providers to claude-council, configures provider API settings, troubleshoots provider connections, and documents the provider script interface. Covers creating provider shell scripts, setting API keys, and validating connectivity. Triggers on "add provider", "new AI agent", "provider not working", "API configuration", or "extend council".