Borrowing it
Nothing to install: this file belongs to snyk-labs/snyk-vulnbench. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/snyk-labs/snyk-vulnbench/main/.agents/skills/benchmark-run/SKILL.mdgit clone --depth 1 https://github.com/snyk-labs/snyk-vulnbenchWrote 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/snyk-labs/snyk-vulnbench/benchmark-run)<a href="https://agentmods.dev/skills/snyk-labs/snyk-vulnbench/benchmark-run"><img src="https://agentmods.dev/badge/skills/snyk-labs/snyk-vulnbench/benchmark-run/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/snyk-labs/snyk-vulnbench/benchmark-run"><img src="https://agentmods.dev/badge/skills/snyk-labs/snyk-vulnbench/benchmark-run.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 | $0.00189 | $0.04187 |
| Opus 5.5 | $0.00076 | $0.01675 |
| Sonnet 5.5 | $0.00038 | $0.00837 |
| Haiku 4.5 | $0.00019 | $0.00419 |
Grade A, and why
benchmark-run 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 27d 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 — 287 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Benchmark Run
Instructions
Turn a natural-language benchmark request into the exact CLI command that runs it, execute it, and report the outcome. No need to memorize task IDs or dig through package.json.
For canonical V2 gateway setup or proxy failures, consult docs/litellm-integration.md before changing runner credentials or provider configuration.
Step 1: Discover available tasks and configs
Read these two sources to build the current inventory:
- Task IDs — list
evals/tasks/*.jsonfilenames. Each filename minus.jsonis the task ID (e.g.js-project-tigerteam-find-vulns). - Config IDs — read
evals/run-configs.json. Each object'sidfield is a config ID. For model configs, also notemcpServersandpromptTemplateId, because they define the available tools and any required tool-use guidance. - Config groups — read
evals/run-config-groups.json. With no selector the CLI usesdefault;vulnbench-v2preserves the original 30-turn DeepSec matrix, whilevulnbench-v2-deepsec-150matches the completed V2 execution.
This step is necessary because tasks and configs change over time — never hard-code the list.
Done when: you have a list of valid task IDs and config IDs.
Step 2: Translate the user's request into CLI arguments
Map the user's natural language to --task, --config, and --category flags.
Task resolution rules:
| User says | Resolves to |
|---|---|
| "js find vulns tigerteam" or "js-project-tigerteam find" | --task js-project-tigerteam-find-vulns |
| "js find vulns tigerteam, shadowfox, ironclad" | --task js-project-tigerteam-find-vulns,js-project-shadowfox-find-vulns,js-project-ironclad-find-vulns |
| "js fix vulns shadowfox" | --task js-project-shadowfox-fix-vulns |
| "llm 1 find" or "llm stardust find" | --task llm-project-stardust-find-vulns |
| "app keystonebank find and fix" | --task app-project-keystonebank-find-vulns,app-project-keystonebank-fix-vulns |
| "python find" | --task python-project-cobalt-find-vulns |
| "all find tasks" or "find vulns" | --category find-vulns (no --task needed) |
| "all fix tasks" | --category fix-vulns |
| "everything" or "all" | omit both --task and --category |
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.
- 27d ago Changed · +61 lines · +7 tokens per session 36d654e22f1a
- 1mo ago First seen · 226 lines · 182 tokens per session scan A a9ebcc6b0997
benchmark-run is a skill published in the GitHub repository snyk-labs/snyk-vulnbench (4 stars, last pushed 14d ago), licensed Apache-2.0. It adds 189 tokens to every session and 4,187 once invoked, about $0.0008 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 2026-08-31.
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".
cua-driver
Use Cua Driver for desktop or browser tasks that are awkward or unavailable through Bash/APIs, or when the user explicitly wants GUI interaction: app testing, visual bug reproduction, form filling, calendar entry, screenshots, and demo recording. Also covers Cua setup; not OpenAI Codex Computer Use or web research.
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.