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 sliamh11/Deus --skill verifygit clone --depth 1 https://github.com/sliamh11/DeusWrote 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/sliamh11/deus/verify)<a href="https://agentmods.dev/skills/sliamh11/deus/verify"><img src="https://agentmods.dev/badge/skills/sliamh11/deus/verify/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/sliamh11/deus/verify"><img src="https://agentmods.dev/badge/skills/sliamh11/deus/verify.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 2 findings, 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 Agent Snooping · line 173 Skill enumerates or reads other installed skills. Access to other skills' SKILL.md files or the skills directory reveals prompt instructions, capabilities, and secrets that should be invisible to peer skills.Fix: Remove all code or instructions that list or read other skills' files or directories. Skills should operate independently; cross-skill access is a privilege escalation.
- medium Agent Snooping · line 174 Skill enumerates or reads other installed skills. Access to other skills' SKILL.md files or the skills directory reveals prompt instructions, capabilities, and secrets that should be invisible to peer skills.Fix: Remove all code or instructions that list or read other skills' files or directories. Skills should operate independently; cross-skill access is a privilege escalation.
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 | $0.00121 | $0.04853 |
| Opus 5 | $0.00060 | $0.02426 |
| Sonnet 5 | $0.00024 | $0.00971 |
| Haiku 4.5 | $0.00012 | $0.00485 |
Grade B, and why
verify scanned grade B with 1 finding 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 7d 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.
Enumerates other installed skillsmediumAgent snooping
Other skills' SKILL.md files reveal prompts, capabilities and secrets that should be invisible to peers.
ls .claude/skills/ How it starts
The opening of the file, as written. The whole thing — 411 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Verify
A verification stands on two legs. Remove either one and it is not a verification.
- Observation at the real surface. The claim is exercised where a user meets it, and what the software actually emitted is captured.
- A controlled contrast. A second capture, from the unchanged state, measured identically — so the difference is attributable to the change and not to the weather.
One leg alone has a name. A treatment capture with no control is a demo. A control/treatment pair measured somewhere other than the real surface is a laboratory result about code nobody runs.
What does not count
Three substitutes, each of which feels like evidence and is not:
- Suites and typecheckers. They re-assert what someone already wrote down. They cannot speak to a claim nobody has encoded yet — which is precisely the claim under verification. Read a test as a specification if it helps; then go exercise the software.
- Calling the changed function directly. Reaching past the public boundary to invoke an internal and printing the result is a test you authored moments ago, graded by its author. The function behaved as written; reading it told you that. Nothing that ships was exercised.
- Narrating the diff. A description of what changed is a claim awaiting verification, never its discharge.
1. Fix the claim, and fix what would refute it
Two sentences, written before anything is measured, and not revised afterwards:
Claim. Under
<condition>, at<surface>,<observable><holds>. Refuted if.<the specific observation that would end this as a failure>.
Worked examples:
Claim: invoked with an empty
--from, the CLI exits non-zero and names the flag in its message. Refuted if: it exits 0, or the message omits--from.
Claim: the twelfth request inside a minute receives 429 carrying
Retry-After. Refuted if: a twelfth request succeeds, or the header is absent or unparseable.
Claim: p95 cold start on this host is at or under 400 ms across 10 trials. Refuted if: p95 exceeds 400 ms, or the trials scatter too widely to place it.
What ships with it
4 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.
- 7d ago First seen · 411 lines · 121 tokens per session scan B bf400bfb64e0
verify is a skill published in the GitHub repository sliamh11/Deus (51 stars, last pushed 5d ago), licensed MIT. It adds 121 tokens to every session and 4,853 once invoked, about $0.0006 per session on Opus 5. A static security scan graded it B with 1 finding (enumerates other installed skills). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-03.
Other skills, from other repositories
red-team-review
Unified adversarial review: v4.3 Strategic Matrix (MTA-004). 7-phase framework: Priors → Rubric → Adversarial Lenses → SWOT/TOWS → MCDA Decision Engine → Blind Spot/Kill Switch → Executive Summary. Absorbs: bias-detector.
consiglieri-protocol
Mandatory pre-flight checklist for high-variance social contracts. Covers the Pryce Test, Exit Test, STFU Clause, Blast Radius Audit, Vibe Veto, and Adult-to-Adult comms rewrite.
diagnostic-first-refactoring
Analyze codebase structure before making changes — the "Surgeon's Scan" pattern.
visual-verify-ui
Wraps the browser tool into a dedicated visual QA testing skill for frontend work.
qa
Systematic QA testing of a web application: diff-aware, tiered, with fix-and-verify loop.
ci-tests
Run the test suite for the current repo, auto-detecting Python (pytest/uv), Node (vitest/pnpm), or Rust (cargo test).