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 ondrej-svec/heart-of-gold-toolkit --skill verifygit clone --depth 1 https://github.com/ondrej-svec/heart-of-gold-toolkitWrote 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/ondrej-svec/heart-of-gold-toolkit/verify)<a href="https://agentmods.dev/skills/ondrej-svec/heart-of-gold-toolkit/verify"><img src="https://agentmods.dev/badge/skills/ondrej-svec/heart-of-gold-toolkit/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/ondrej-svec/heart-of-gold-toolkit/verify"><img src="https://agentmods.dev/badge/skills/ondrej-svec/heart-of-gold-toolkit/verify.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00055 | $0.01730 |
| Opus 5 | $0.00028 | $0.00865 |
| Sonnet 5 | $0.00011 | $0.00346 |
| Haiku 4.5 | $0.00006 | $0.00173 |
Grade A, and why
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 11d 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 — 140 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Verify (walk the contract before claiming done)
A pre-Stop verification pass that walks the active evidence contract claim-by-claim, gathers the evidence each claim requires, and reports pass/fail per claim. This is the agent's tool for proving readiness, not a backdoor around the Stop hook.
If a claim's evidence is present, the Stop hook will let the agent say the corresponding thing. If it's missing, this skill says so plainly so the agent fixes the work, not the message.
Boundaries
This skill MAY: read the contract, run commands listed in the contract's evidence kinds (cargo test, pytest, SELECT …, gitleaks, git diff), write a verify report to .quellis/contracts/<task-id>-verify.md, after a full-PASS report, prompt the user to mark the contract completed and — only on explicit yes — flip status = "completed" and clear the active.toml pointer.
This skill MAY NOT: edit code, modify the contract to make claims pass, auto-complete the contract without the user's explicit yes, run destructive commands (drops, deletes, force pushes).
Verify gathers evidence. It does not generate evidence. If tests fail, the verify report says they fail. The agent's response is to fix the code, not rerun verify hoping for different output.
Common Rationalizations
| Shortcut | Why it fails | The cost |
|---|---|---|
| "Tweak the claim_regex so my claim no longer matches" | Defeats the contract. The gate exists because the claim is dangerous to assert without evidence. | Silent regressions; the next agent inherits a lying contract. |
| "Auto-mark completed on PASS — skip the prompt" | A verify run that passes can still be wrong (false-positive evidence: test ran but tested the wrong thing). The human-yes is the last guard. | Premature completion buries claims that should have triggered a second look. |
| "Run verify until it passes" | If a claim fails, the answer is to make the code prove the claim — not to rerun the same gather. | Wasted cycles; underlying issue persists. |
| "Skip claims that don't have evidence yet" | Skipping is the contract's job, not yours. Stop's evidence-search already suppresses claims with present evidence. | Verify report disagrees with Stop hook; agent confused about which to trust. |
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.
- 11d ago First seen · 140 lines · 55 tokens per session scan A a3f77d648b9f
verify is a skill published in the GitHub repository ondrej-svec/heart-of-gold-toolkit (19 stars, last pushed 22d ago), licensed MIT. It adds 55 tokens to every session and 1,730 once invoked, about $0.0003 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 2026-08-30.
Other skills, from other repositories
research-engineer
An uncompromising Academic Research Engineer. Operates with absolute scientific rigor, objective criticism, and zero flair. Focuses on theoretical correctness, formal verification, and optimal implementation across any required technology.
tika-eval-compare
Compare extracts from two Tika builds over a corpus to detect regressions in content, encoding, exceptions, and embedded-document handling. Use for "compare before/after extracts", "eval this change against the corpus".
neuron-evaluation-engineer
Create and run AI evaluations with datasets, assertions, and output drivers in Neuron AI. Use this skill whenever the user mentions evaluation, testing AI systems, creating evaluators, dataset-driven testing, assertion-based validation, or wants to measure AI system performance. Also trigger for tasks involving…
jetson-validate-image
Use after jetson-flash-image to run static BSP checks, on-target smoke/regression tests on a flashed DUT, or both. Not for build or flash steps. Triggers: validate bsp, on-target validation.
atmos-validation
Validate Atmos projects, components, arbitrary JSON Schema inputs, EditorConfig, and GitHub Actions; use affected-file selection and native CI annotations.
skill-benchmark
Benchmark AI skill effectiveness by measuring implementation quality against legacy constraints.