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 build-with-dhiraj/ai-workflow-framework-portability-kit --skill llm-councilgit clone --depth 1 https://github.com/build-with-dhiraj/ai-workflow-framework-portability-kitWrote 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/build-with-dhiraj/ai-workflow-framework-portability-kit/llm-council)<a href="https://agentmods.dev/skills/build-with-dhiraj/ai-workflow-framework-portability-kit/llm-council"><img src="https://agentmods.dev/badge/skills/build-with-dhiraj/ai-workflow-framework-portability-kit/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/build-with-dhiraj/ai-workflow-framework-portability-kit/llm-council"><img src="https://agentmods.dev/badge/skills/build-with-dhiraj/ai-workflow-framework-portability-kit/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 | $0.00202 | $0.00740 |
| Opus 5.5 | $0.00081 | $0.00296 |
| Sonnet 5.5 | $0.00040 | $0.00148 |
| Haiku 4.5 | $0.00020 | $0.00074 |
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 1mo 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 — 51 lines — stays where its author put it; the contents beside it link to each section on GitHub.
LLM Council
Five advisors answer independently, then review each other blind, then a chairman synthesizes. One model, five thinking lenses, plus a peer-review round that catches what any single pass misses.
Run it when being wrong is expensive. Skip it when there's one right answer, or when the user wants something made rather than judged.
Workflow
1. Enrich, then frame. Glob/Read for context that would make advice specific instead
of generic: CLAUDE.md, any memory/ folder, files the user referenced, prior
council-transcript-*.md in the cwd. Cap it at 2-3 files. Then write one neutral framed
question carrying: the decision, the user's context, the workspace context (stage,
audience, constraints, numbers), and what's at stake. No opinion, no steering. If the
question is too vague to frame, ask exactly one clarifying question, then proceed.
2. Convene. Dispatch all 5 advisors in a single message so they run in parallel. Prompts in ADVISORS.md. 150-300 words each.
3. Peer review. Shuffle the 5 responses into A-E (record the mapping). Dispatch 5 reviewers in parallel, each seeing all five anonymized responses. Under 200 words each.
4. Chairman. One agent gets the framed question, the de-anonymized responses, and all 5 reviews. Fixed five-section verdict.
5. Render. Write the session to JSON, then:
python3 ~/.claude/skills/llm-council/scripts/render.py session.json .
It emits council-report-<timestamp>.html and council-transcript-<timestamp>.md in the
cwd and prints both paths. Send the HTML to the user with SendUserFile
(display: "render"). JSON schema is documented at the top of the script.
Rules
- Parallel, always. Sequential dispatch lets earlier responses bleed into later ones.
- Anonymize for review. Named advisors get deferred to by reputation, not merit.
- Advisors do not hedge. Each leans fully into its lens. Balance is the chairman's job.
- The chairman may side with the minority when the dissenter's reasoning is strongest, and should say why.
- Don't council trivia. If there's one right answer, just answer it.
What ships with it
2 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.
- 1mo ago First seen · 51 lines · 202 tokens per session scan A 611fce4e472f
llm-council is a skill published in the GitHub repository build-with-dhiraj/ai-workflow-framework-portability-kit (5 stars, last pushed 21d ago), licensed MIT. It adds 202 tokens to every session and 740 once invoked, about $0.0008 per session on Opus 5.5. 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 2026-09-21.
Other skills, from other repositories
codeql
Run CodeQL database creation and security queries, add data-extension models, or process CodeQL SARIF. Use when CodeQL is explicitly requested; use security-review for a broader manual security review.
sharp-edges
Identify error-prone APIs, dangerous configuration, and footgun designs that enable security mistakes. Use for API ergonomics, misuse resistance, secure defaults, or pit-of-success reviews; use code-review for general defects.
huggingface-vision-trainer
Train object-detection, image-classification, or SAM segmentation models on Hugging Face Jobs. Use for vision fine-tuning and evaluation; use huggingface-llm-trainer for language models.
code-quality
Agents should invoke this skill for code reviews, linting/formatting setup, maintainability checks, complexity concerns, warning cleanup, coding standards, or quality gates in Rust, TypeScript, Python, shell, and mixed repos.
code-reviewer
Review code for bugs, logic errors, security vulnerabilities, code quality issues, and adherence to project conventions, using confidence-based filtering to report only high-priority issues that truly matter. Use this skill when reviewing a small set of changes locally (such as unstaged diff), when dispatched as a…
differential-review
Performs security-focused differential review of code changes (PRs, commits, diffs). Adapts analysis depth to codebase size, uses git history for context, calculates blast radius, checks test coverage, and generates comprehensive markdown reports. Automatically detects and prevents security regressions.