Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add gcpdev/llm-council-skill --skill llm-councilgit clone --depth 1 https://github.com/gcpdev/llm-council-skillWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/gcpdev/llm-council-skill/llm-council)<a href="https://agentmods.dev/skills/gcpdev/llm-council-skill/llm-council"><img src="https://agentmods.dev/badge/skills/gcpdev/llm-council-skill/llm-council/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/gcpdev/llm-council-skill/llm-council"><img src="https://agentmods.dev/badge/skills/gcpdev/llm-council-skill/llm-council.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00082 | $0.00988 |
| Opus 5 | $0.00041 | $0.00494 |
| Sonnet 5 | $0.00016 | $0.00198 |
| Haiku 4.5 | $0.00008 | $0.00099 |
Grade A, and why
llm-council scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 10d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 89 lines — stays where its author put it; the contents beside it link to each section on GitHub.
LLM Council
Consult multiple AI models (ChatGPT and Gemini) for their perspectives before presenting implementation plans to users.
Workflow
When user requests consultation with other AI models, use phrases like:
- "Consult with ChatGPT and Gemini about..."
- "Ask other AI models what they think about..."
- "Get perspectives from the council on..."
- "Consult the LLM council: [your question]"
Process:
- Query external LLMs: Run
scripts/query_llms.pywith the user's prompt to get perspectives from both ChatGPT and Gemini - Analyze responses: Review what each model suggests, identifying valuable insights, alternative approaches, and potential concerns
- Synthesize plan: Create an implementation plan that incorporates the best ideas from all three models (Claude's own analysis + ChatGPT + Gemini)
- Present to user: Show the final plan along with a brief summary of key contributions from each model
Setup Requirements
The skill requires API keys and optional model configuration stored in a .env file in the working directory:
OPENAI_API_KEY=sk-...
GEMINI_API_KEY=...
# Optional: Specify which models to use (defaults shown below)
OPENAI_MODEL=gpt-5-nano
GEMINI_MODEL=gemini-3-flash-preview
Default Models:
- ChatGPT:
gpt-5-nano(fastest, most cost-efficient - $0.05/1M input, $0.40/1M output) - Gemini:
gemini-3-flash-preview(balanced speed and intelligence)
Upgrade Options for Better Collaboration:
OpenAI models (ordered by capability and cost):
gpt-5-nano- Fastest, most cost-efficient ($0.05/1M in, $0.40/1M out) - DEFAULTgpt-5-mini- Faster, cost-efficient for well-defined tasks ($0.25/1M in, $2.00/1M out)gpt-5.2- Best for coding and agentic tasks ($1.75/1M in, $14.00/1M out)gpt-5.2-pro- Smarter, more precise for complex problems ($21.00/1M in, $168.00/1M out)
All models support reasoning tokens, 400K context window, and image input.
Gemini models (ordered by capability):
gemini-2.5-flash-lite- Ultra-fast, optimized for throughputgemini-2.5-flash- Best price-performance, large-scale processinggemini-3-flash-preview- Balanced speed and frontier intelligence (default)gemini-3-pro-preview- Most intelligent multimodal model, best for complex reasoning
What ships with it
3 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 10d ago First seen · 89 lines · 82 tokens per session scan A 7daba585e6fe
llm-council is a skill published in the GitHub repository gcpdev/llm-council-skill (438 stars, last pushed 8mo ago), licensed MIT. It adds 82 tokens to every session and 988 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
recursive-decomposition
Handle tasks that exceed the context window by decomposing them: size and filter the input, chunk it, run recursive sub-agents on independent parts, verify on small windows, and synthesise programmatically, following the Recursive Language Models (RLM) research by Zhang, Kraska and Khattab (2025). Use when a task…
claude-nextsession
Schedule a set of instructions to run automatically once the current Claude Code usage session resets, using the local rate-limit cache (no API key needed). Trigger: "postpone this to next session", "run this once my usage resets", "queue this for later / after the reset", /claude-nextsession.
security-audit
Conduct a security-led audit of code, systems, APIs, infrastructure, dependencies, and AI workflows with threat modeling and exploitability-backed remediation. Use when security risk is the primary concern. Do not use for a generic code review, database design, or non-security debugging.
architecture-planning
Plan the future architecture of a new or materially changing system before implementation. Use for greenfield products, planned platform changes, and technical blueprints spanning components, data, APIs, deployment, and delivery trade-offs. Do not use to audit an existing repository, review a PR, or diagnose an…
code-review
Review a specific implementation, change set, PR, function, or service for broad correctness and production-readiness findings. Use when code-quality is the primary concern. Do not use when security, performance, architecture, dependencies, or tests are the exclusive requested discipline.
debugging
Investigate and safely fix an active, observable software failure, error, regression, or unexpected behavior using reproduction and evidence. Use for the proximate technical cause and remediation. Do not use for a systemic incident postmortem or general performance tuning without a failure.