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 goiltpatpat/patpat --skill patpat-verifygit clone --depth 1 https://github.com/goiltpatpat/patpatWrote 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/goiltpatpat/patpat/patpat-verify)<a href="https://agentmods.dev/skills/goiltpatpat/patpat/patpat-verify"><img src="https://agentmods.dev/badge/skills/goiltpatpat/patpat/patpat-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/goiltpatpat/patpat/patpat-verify"><img src="https://agentmods.dev/badge/skills/goiltpatpat/patpat/patpat-verify.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 | $0.00041 | $0.00800 |
| Opus 5 | $0.00020 | $0.00400 |
| Sonnet 5 | $0.00008 | $0.00160 |
| Haiku 4.5 | $0.00004 | $0.00080 |
Grade A, and why
patpat-verify 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.
How it starts
The opening of the file, as written. The whole thing — 48 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Patpat Verify
For a bounded local proof, use this skill and its relevant references. Read the operating protocol in full for uncertainty, security, auth, billing, secrets, architecture, cross-cutting work, or delivery intent. Do not load the router or reread unchanged instructions already loaded in this session.
Read proof over proxy and preserve safety.
Build the proof
- Reconcile the requested outcomes with the proof contract using proof over proxy; name missing requirements before running checks.
- Identify the authoritative artifact or user surface.
- Capture a baseline or reproduce the prior state when relevant.
- Run the smallest targeted automated check.
- For behavioral claims, exercise the changed logic through the authoritative interface under representative conditions when safe and practical. For static claims, use the strongest deterministic check on the authoritative artifact.
- Observe visible output and material side effects directly.
- Inspect the final diff and version-control state for unrelated changes.
- Record the command or action, observed result, cleanup, and limitation for each material claim; do not mark the whole request verified with uncovered outcomes.
For every test or evaluator used as evidence, apply the behavioral check from proof over proxy: a mock-call assertion, copied constant, self-referential expected value, or fixture-only agreement cannot prove user-observed behavior. Replace it with a concrete input and literal expected output or side effect, or classify the claim as static instead.
Claim-adaptive verification and proxy rejection
Match verification depth to the claim and risk:
- Behavioral claims: Vary material inputs, relevant error branches, or state transitions when safe and practical. Call the subject through the authoritative interface and assert the literal observed result or material effect. A mock interaction, call count, truthiness check, copied constant, or clean compilation does not prove the behavior by itself.
- Static claims: Use deterministic structure, schema, type, or content checks when that is the authoritative surface. Do not manufacture runtime theater for a non-runtime claim.
- Proxy evidence: Fixtures and mocks may isolate a contract, but they do not replace the real system when the claim concerns that system. Reject narrative summaries, hardcoded fixture answers, and clean compilation as sole behavioral proof.
- Fresh binding: Bind evidence to the exact candidate revision or a reproducible working-tree snapshot, plus material inputs, environment, and oracle. Require the committed head for commit-, push-, PR-, or delivery-bound claims. Reject stale, cached, or transferred logs.
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 Changed · +2 lines 2565a897ce3f
- 3d ago Changed 14b8652dc452
- 10d ago First seen · 46 lines · 41 tokens per session scan A 689b1a725944
patpat-verify is a skill published in the GitHub repository goiltpatpat/patpat (1 stars, last pushed yesterday), licensed MIT. It adds 41 tokens to every session and 800 once invoked, about $0.0002 per session on Opus 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 2026-08-31.
Other skills, from other repositories
gentle-ai-bench
Trigger: bench, journey, journeys, driven mode, gentle-ai-bench, journey corpus, j-numbers, bench axis. Author and verify gentle-ai bench journeys; go test ./bench never proves driven execution.
go-testing
Trigger: Go tests, go test coverage, Bubbletea teatest, golden files. Apply focused Go testing patterns.
coding-agents-hooks-authoring
To author, register, and test Rosetta hooks, add a SemanticKind, or debug a hook that won't fire.
ijfw-verify
Use when about to claim completion: 'done', 'fix complete', 'tests pass', 'build succeeded', 'shipped', 'no regressions', 'ready to merge', 'ready to ship'. Iron Law gate requiring fresh verification evidence in the same message as the claim; wires into runtime (verification-gate.js + the ijfwstate MCP tool…
ijfw-tdd
RED-GREEN-REFACTOR enforcement. Use when implementing a feature or bugfix before writing implementation code. Trigger: tdd, test first, red green refactor, /ijfw-tdd.
api-aqa-flow
Workflow for backend API test automation: TMS / Issue Tracker test cases → automated API tests, HITL-gated.