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 CSOAI-ORG/councilof-ai --skill verify-cardgit clone --depth 1 https://github.com/CSOAI-ORG/councilof-aiWrote 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/csoai-org/councilof-ai/verify-card)<a href="https://agentmods.dev/skills/csoai-org/councilof-ai/verify-card"><img src="https://agentmods.dev/badge/skills/csoai-org/councilof-ai/verify-card/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/csoai-org/councilof-ai/verify-card"><img src="https://agentmods.dev/badge/skills/csoai-org/councilof-ai/verify-card.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 | $0.00060 | $0.00520 |
| Opus 5.5 | $0.00024 | $0.00208 |
| Sonnet 5.5 | $0.00012 | $0.00104 |
| Haiku 4.5 | $0.00006 | $0.00052 |
Grade A, and why
verify-card 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 today.
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.
What it actually says
Verify a measurement card
A card is a gspc.measurement-card. Verification recomputes the card id from the canonical body and
checks the Ed25519 signature under a published pinned key in the DID document
did:web:csoai.org.
The rule
- Recompute the card id from the canonical body bytes.
- Check the Ed25519 signature under a key that is pinned in the verifier and published in the DID
document:
#card-attestation-1,#card-attestation-2, plus the board key. - A card that is internally consistent but signed with a freshly generated key is INVALID, not valid. Anyone can alter a body and sign it with a key they just made.
- A card naming an inline key outside the pinned set is INVALID. A card naming a DID key reference outside the pinned set is UNCHECKABLE.
Three verdicts, never two: VALID, INVALID (with the reason), UNCHECKABLE. "Could not check" is a
different claim from "forged".
How
Prefer MCP verify_card with { "card": <object | JSON string | https://councilof.ai/... URL | 64-hex id> }.
Human UI: https://councilof.ai/verify — the browser recomputes locally; nothing needs to be sent to Council of AI.
Output
verdict: VALID | INVALID | UNCHECKABLE
reason: <one line>
id: <recomputed or stated>
key: did:web:csoai.org#card-attestation-1
Verdict first. Only if VALID, quote the measured axes and the empty slots exactly as the card states them.
UNCHECKABLE when the payload is truncated, the schema is unknown, or a URL was given whose origin is not councilof.ai / csoai.org.
Forbidden
- Summarising the score before the verdict
- Turning UNCHECKABLE into FAIL, or INVALID into "probably fine"
- Verifying a card the user has not given you
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.
- today First seen · 52 lines · 60 tokens per session scan A 1af5e4ad3d5f
verify-card is a skill published in the GitHub repository CSOAI-ORG/councilof-ai (3 stars, last pushed today), licensed MIT. It adds 60 tokens to every session and 520 once invoked, about $0.0002 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 2026-10-27.
Other skills, from other repositories
complyedge-eu-ai-act-check-and-cite
Check this prompt or output against EU AI Act Article 5 and Article 50 rules and cite the article, via the ComplyEdge TrustLint MCP. Use when the user wants agent I/O checked and cited, not system classification, FRIA forms, or C2PA watermarking.
governance-setup
Configure PromptSpeak governance for a new project. Use when first setting up PromptSpeak or when the user wants to customize governance policies.
code-executor-mcp
Secure code execution in sandboxed environments. Run Python, JavaScript, and shell commands safely with tier-based permissions.
memory-search-mcp
Long-term memory and context retrieval for AI agents. Semantic search across conversation history with care-weighted relevance.
web-research-mcp
Intelligent web research with source validation. Automated research synthesis with citation tracking and fact verification.
x402-payments
Work with Server OS micropayment (x402) capabilities for agents.