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 Skull-boy/agent-contracts --skill scyvera-governancegit clone --depth 1 https://github.com/Skull-boy/agent-contractsWrote 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/skull-boy/agent-contracts/scyvera-governance)<a href="https://agentmods.dev/skills/skull-boy/agent-contracts/scyvera-governance"><img src="https://agentmods.dev/badge/skills/skull-boy/agent-contracts/scyvera-governance/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/skull-boy/agent-contracts/scyvera-governance"><img src="https://agentmods.dev/badge/skills/skull-boy/agent-contracts/scyvera-governance.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 | $0.00074 | $0.01483 |
| Opus 5.5 | $0.00030 | $0.00593 |
| Sonnet 5.5 | $0.00015 | $0.00297 |
| Haiku 4.5 | $0.00007 | $0.00148 |
Grade A, and why
scyvera-governance 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 4d 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 — 112 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Scyvera governance
Scyvera is a deterministic governance CLI. This skill teaches you when to run it, which command to run, and how to read what comes back. It does not give you governance authority.
Design principle: knowledge, not powers
A skill must never give an agent governance POWERS, only governance KNOWLEDGE. You run Scyvera's tools; you do not become the judge. The engines decide PASS/FAIL from schemas, rules, and adapters. Your job is to invoke the right command, preserve its output, and report its exit code. Do not execute scanned code, do not edit a finding to change its meaning, and do not overrule an engine verdict with your own opinion.
When to invoke
- Asked to audit, scan, or assess blast radius: run
scyvera audit. - Asked whether a contract is valid or semantically sane: run
scyvera validate, thenscyvera lint. - Asked whether an implementation conforms, or to check before a consequential action: run
scyvera check. - Asked to prove an installed Scyvera is intact: run
scyvera self-check. - Asked to score a trace-driven case: run
scyvera eval. - Asked for a starting contract: run
scyvera init, then validate.
The three NEVERs
- NEVER fabricate findings. Report only what the command printed. If it failed, report the failure.
- NEVER lower a finding's severity. Copy the engine's severity and rule ID unchanged. Do not re-rank, suppress, or explain away a finding.
- NEVER declare PASS without running the command. "It looks fine" is not a verdict. A PASS claim requires a real run and exit code 0.
Command map
Exit codes are the contract: 0 success/pass (warnings may still print), 1 a
finding or verdict failure, 2 usage, IO, or unverifiable input. Never treat
stderr as a pass.
scyvera init [output] — scaffold
Writes the starter contract (default contract.yaml). Exit 0 created, 1 target exists or cancelled.
$ scyvera init contract.yaml --name "Demo Agent"
CREATED contract.yaml (spec version 1.1) # exit 0
What ships with it
5 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.
- 4d ago First seen · 112 lines · 74 tokens per session scan A 3cbf326921ba
scyvera-governance is a skill published in the GitHub repository Skull-boy/agent-contracts (25 stars, last pushed yesterday), licensed MIT. It adds 74 tokens to every session and 1,483 once invoked, about $0.0003 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 2026-11-14.
scyvera-governance compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| scyvera-governanceSkull-boy/agent-contracts | Skill | 25 | yesterday | MIT | A | 74 | Claude Code, Codex |
| build-workflowneurawork-git/n8n-autopilot | Skill | 18 | 16d ago | MIT | A | 85 | Claude Code |
| n8nac-cheatsheetneurawork-git/n8n-autopilot | Skill | 18 | 16d ago | MIT | A | 86 | Claude Code |
| deployneurawork-git/n8n-autopilot | Skill | 18 | 16d ago | MIT | A | 54 | Claude Code |
| n8n-code-javascriptneurawork-git/n8n-autopilot | Skill | 18 | 16d ago | MIT | A | 63 | Claude Code |
| n8n-orchestration-patternsneurawork-git/n8n-autopilot | Skill | 18 | 16d ago | MIT | A | 84 | Claude Code |
| session-envneurawork-git/n8n-autopilot | Skill | 18 | 16d ago | MIT | A | 108 | Claude Code |