Borrowing it
Nothing to install: this file belongs to julien777z/code-review-action. 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/julien777z/code-review-action/main/.agents/skills/skill-gauntlet/SKILL.mdgit clone --depth 1 https://github.com/julien777z/code-review-actionWrote 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/julien777z/code-review-action/skill-gauntlet)<a href="https://agentmods.dev/skills/julien777z/code-review-action/skill-gauntlet"><img src="https://agentmods.dev/badge/skills/julien777z/code-review-action/skill-gauntlet.svg" alt="Measured on agentmods" 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 | $0.00071 | $0.00894 |
| Opus 5 | $0.00036 | $0.00447 |
| Sonnet 5 | $0.00014 | $0.00179 |
| Haiku 4.5 | $0.00007 | $0.00089 |
Grade A, and why
skill-gauntlet 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 8d 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.
This is a copy
100% identical to skill-gauntlet — 0 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.
How it starts
The opening of the file, as written. The whole thing — 67 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Skill Gauntlet
Build evidence that a skill adds durable value to the current model. Preserve originals, keep evaluation roles independent, and prefer honest retirement over a manufactured improvement.
Resources
- Read references/protocol.md completely before starting a run.
- Use
scripts/gauntlet.pyto initialize resumable run state, snapshot skills, verify snapshot integrity, and serve the view-only dashboard. - Use assets/dashboard.html only through that utility so sealed artifacts are never served.
Workflow
- Discover every installed skill across all project, user, system, plugin, and provider scopes visible to the current harness. Resolve symlinks and precedence, but retain each installation record even when identical content is deduplicated.
- Initialize a run under
~/.agents/skill-gauntlet/runs/<run-id>/. Snapshot every complete skill directory before analysis, verify its digest, and mark whether its installed source is editable. - Record each skill's purpose, origin, dependencies, overlaps, precedence, and editability in the public state. Start the loopback dashboard, open it with the available system browser, give the user its URL in chat, present the inventory in chat, and wait for the user's selection. Never put selection controls in the dashboard or treat dashboard activity as approval.
- After selection, run the complete protocol independently for each selected skill. Do not ask the user to choose experiments, interpret results, approve revisions, or direct iterations.
- Pause only for a genuine external blocker, an action that can affect live systems or data, unavailable isolation needed to protect evaluation integrity, or unverifiable model identity.
- Keep public dashboard state current after each completed phase. Persist sealed tasks and
evaluation packets only in the run's non-served
sealed/directory and never disclose them to builders or contestants before final evaluation. - Keep the lead agent as the only state writer. Serialize every utility mutation after receiving
independent-agent results; never let parallel agents call
snapshotorupdateconcurrently. - Provisionally install a candidate only after it satisfies the frozen iteration acceptance gate. Install a writable user-scope override when a managed source cannot be edited and precedence rules support an override. Otherwise report the installation blocker.
- Run the final held-out acceptance gate after installation. If it fails, restore the verified original snapshot, record the rollback, archive and replace the contaminated held-out set, and resume iteration.
- Finish only when every selected skill is honestly green as an upgrade or
Green — retire.
What ships with it
3 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.
- 8d ago First seen · 67 lines · 71 tokens per session scan A f926a43a21bd
skill-gauntlet is a skill published in the GitHub repository julien777z/code-review-action (2 stars, last pushed 3d ago), licensed MIT. It adds 71 tokens to every session and 894 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it A with 0 findings. It is 100% identical to skill-gauntlet, differing in 0 lines, and is treated as a copy.
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.