Getting it into your agent
It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.
git clone --depth 1 https://github.com/mishahanin/heading-osnpx agentmods add skills/mishahanin/heading-os/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/mishahanin/heading-os/council)<a href="https://agentmods.dev/skills/mishahanin/heading-os/council"><img src="https://agentmods.dev/badge/skills/mishahanin/heading-os/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/mishahanin/heading-os/council"><img src="https://agentmods.dev/badge/skills/mishahanin/heading-os/council.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 1 finding, up to medium
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- medium analysis-evasion · line 1 Suspicious Unicode normalization or mixed-script contentFix: Review the flagged content for security risks. Ensure no credentials, secrets, or sensitive data are exposed.
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 | $0.00157 | $0.04285 |
| Opus 5.5 | $0.00063 | $0.01714 |
| Sonnet 5.5 | $0.00031 | $0.00857 |
| Haiku 4.5 | $0.00016 | $0.00428 |
Grade A, and why
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 — 334 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Council - Independent Second Opinions (Kimi k3 by default; Gemini, Grok and ChatGPT opt-in)
Independent second opinions dispatched in parallel. Use when:
- The user wants fresh views on a hard call (independent mode)
- The user wants a draft stress-tested before it ships (critique mode)
This skill is distinct from /deep-think (Claude reasoning harder, alone) and /odin (Claude + the curated knowledge brain). The unique value is a model with a different training pedigree, different RLHF, and different failure modes reaching the question independently. Agreement is stronger evidence, and disagreement is itself information. --all calls every voice on the roster when breadth matters more than latency.
Phase 0 - Determine mode
Read the user's request.
CRITIQUE mode if any of:
- The user passed
--critique - The user said "stress-test this", "critique this draft", "find flaws in", "what's wrong with this", "review this draft"
- The user pasted a draft (proposal, message, claim) and asked for review
INDEPENDENT mode otherwise (default).
Phase 1 - Gather inputs
For INDEPENDENT mode, prepare:
question: the user's question, cleaned and concretecontext: facts the user has shared in this conversation that bear on the question. Important: include the FACTS (numbers, names, dates, constraints), NOT Claude's reasoning or proposed answer. The whole point of independent mode is to give every outside voice a clean slate.
For CRITIQUE mode, prepare:
draft: the exact draft text to critique. If the user passed--critique 'text'inline, the draft is the quoted text immediately following the flag. If the user pasted a draft in the message body without the flag, use that body verbatim.context: why the draft was produced (audience, goal, constraints). In critique mode, Claude's reasoning IS appropriate to include. The outside voices are being asked to stress-test it.
Optional — to give Claude's view a distinct lens from the outside models, pull 2-5 methods from reference/elicitation-methods.md (python scripts/elicit.py list --category collaboration; categories collaboration/research/framing, e.g. Steelmanning, Reframe the Question). Skip when the question is already well-framed.
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.
- 1mo ago Changed · +6 lines 8e4461151829
- 1mo ago Changed · -10 lines · +14 tokens per session ba9ae10cece5
- 2mo ago First seen · 338 lines · 143 tokens per session scan A 09641e74a979
council is a skill published in the GitHub repository mishahanin/heading-os (11 stars, last pushed 1mo ago), licensed Apache-2.0. It adds 157 tokens to every session and 4,285 once invoked, about $0.0006 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 2026-08-30.
Other skills, from other repositories
dcf-model
Build discounted cash flow valuation workbooks in Excel.
mle-workflow
Production machine-learning engineering workflow for data contracts, reproducible training, model evaluation, deployment, monitoring, and rollback. Use when building, reviewing, or hardening ML systems beyond one-off notebooks.
ascii-video
ASCII video: convert video/audio to colored ASCII MP4/GIF.
grounded-citations
Ground answers and documents in cited, verifiable sources.
hermes-agent-skill-authoring
Author in-repo SKILL.md files: frontmatter and structure.
systematic-debugging
4-phase root cause debugging: understand bugs before fixing.