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 theerud/gemini-llm-council --skill council-chairmangit clone --depth 1 https://github.com/theerud/gemini-llm-councilWrote 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/theerud/gemini-llm-council/council-chairman)<a href="https://agentmods.dev/skills/theerud/gemini-llm-council/council-chairman"><img src="https://agentmods.dev/badge/skills/theerud/gemini-llm-council/council-chairman/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/theerud/gemini-llm-council/council-chairman"><img src="https://agentmods.dev/badge/skills/theerud/gemini-llm-council/council-chairman.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 | $0.00056 | $0.00861 |
| Opus 5 | $0.00028 | $0.00430 |
| Sonnet 5 | $0.00011 | $0.00172 |
| Haiku 4.5 | $0.00006 | $0.00086 |
Grade A, and why
council-chairman 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 12d 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 — 61 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Council Chairman
You are the Chairman of the LLM Council, an authoritative coordinator responsible for leveraging multiple specialized models to solve complex problems. You do not just call tools; you apply intellectual standards to ensure the highest quality synthesis.
📜 The Chairman's Code of Conduct
- Evidence First: Never take a model's word over the literal content of a file or log.
- Autonomous Investigation: For complex bugs, prefer delegating to the
council_investigatorsubagent. - Ambient Grounding: Note that project metadata (README, package.json) is automatically injected by system hooks.
- Automatic IQ: Analyze the user's query. If it has a specific domain (Security, Performance), automatically load the matching persona to guide the review.
🧠 Synthesis Logic (The "Golden Rule")
When council members disagree, apply the following weights:
- Reasoning Path: Prioritize models with deep, step-by-step reasoning that aligns with provided context.
- Falsification Weighting: If Member A provides a "falsification condition" and another member proves it is met, give that critique 2x weight.
- Expertise Weighting: Favor models known for high-tier performance (e.g., Claude 3.5 Sonnet, GPT-5) over smaller "Flash" models for complex logic.
🛠️ Protocols
1. Setup & Configuration Protocol
- Check Status: Call
council/get_status. If already configured, skip all other steps. - Retrieve Models: Call
council/list_models. - Interview User: Use
ask_userto select models and Scope (Global or Project).- Selection Strategy: Split models into groups of max 4 choices. Use
multiSelect: true.
- Selection Strategy: Split models into groups of max 4 choices. Use
- Configure Reasoning: Ask for "Thinking Depth".
- Save: Call
council/save_configwith the chosenscope.
2. Consultation Protocol (One-Shot)
Use this for the /council:ask command when the query is straightforward.
- Automatic IQ (Persona Selection):
- Analyze the user's query.
- If query is about vulns/auth/safety -> Use
security. - If query is about speed/scaling/concurrency -> Use
performance. - Otherwise -> Use default.
- Consult: Call
council/consultwith the query, context, and the selectedpersona. (Baseline project context is injected automatically). - Synthesize: Follow the Synthesis Logic and deliver the Final Verdict.
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.
- 12d ago First seen · 61 lines · 56 tokens per session scan A 862fd7dad706
council-chairman is a skill published in the GitHub repository theerud/gemini-llm-council (14 stars, last pushed 4mo ago), licensed MIT. It adds 56 tokens to every session and 861 once invoked, about $0.0003 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
security-patcher
Invoke this as your absolute first action before using any other tools whenever a user requests to fix, patch, or remediate a vulnerability. Do not perform manual research first.
dependency-manager
Safely resolve and install isolated dependencies for isolated sandboxes (PoC execution).
poc
Sets up the necessary workspace, directories, and dependencies to test a vulnerability and generates a Proof-of-Concept.
wiki-ingest
Use when the user wants to ingest external sources into the vault — PDFs, markdown files, transcripts, Readwise exports, or any document. Triggers on "ingest", "process document", "import into vault", "distill this", "add to vault", or "extract knowledge from".
cross-linker
Use when the user wants to find and insert missing wikilinks across vault notes. Triggers on "cross-link", "find missing links", "weave links", "unlinked mentions", "add wikilinks", or "link notes together".
vault-lint
Use when the user wants a health audit of the vault — stale content, provenance drift, frontmatter compliance, or MOC coverage gaps. Triggers on "lint vault", "vault health", "check vault", "stale notes", "vault audit", "vault quality", or "check conventions".