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 agentmods add skills/sanjoy1234/gauntlex/validatenpx skills add sanjoy1234/gauntlex --skill validategit clone --depth 1 https://github.com/sanjoy1234/gauntlexWrote 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/sanjoy1234/gauntlex/validate)<a href="https://agentmods.dev/skills/sanjoy1234/gauntlex/validate"><img src="https://agentmods.dev/badge/skills/sanjoy1234/gauntlex/validate.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 | $0.00048 | $0.00409 |
| Opus 5 | $0.00024 | $0.00204 |
| Sonnet 5 | $0.00010 | $0.00082 |
| Haiku 4.5 | $0.00005 | $0.00041 |
Grade A, and why
gauntlex:validate scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 4d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
What it actually says
/gauntlex:validate
Validates the GAUNTLEX environment and spec WITHOUT firing any real attacks.
Usage
/gauntlex:validate [--spec <file>]
Execution
gauntlex validate --spec "$SPEC_FILE"
What It Checks
- AVF gate: Breaker must find >75% of seeded vulnerabilities in the 5 golden CVE fixtures
- Model reachability: GAUNTLEX never assumes a provider — it checks whichever provider was explicitly configured (Anthropic, OpenRouter, HuggingFace, OpenAI-compatible, or local Ollama, set via
gauntlex setup,MODEL_PROVIDER, or.gauntlex.yml). If nothing is configured, this fails with "not configured — rungauntlex setup" - ChromaDB: Knowledge Forge writable at
.gauntlex/forge/ - Playbook resolution: All domains listed in
.gauntlex.ymlhave corresponding YAML files - Spec parseable: The spec file (if provided) can be read and is non-empty
Output to User
Report each check as PASS/FAIL with a reason, using the exact Detail text from the command output for the model check — do not assume or guess a provider. If any FAIL:
- Explain the specific failure
- Give the fix command shown in the output (
gauntlex setupif no provider is configured at all; otherwise the provider-specific fix, e.g.ollama pull llama3.1:8bonly if Ollama is the provider actually configured)
If all pass: "Environment validated — ready to run."
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.
- 4d ago First seen · 41 lines · 48 tokens per session scan A eff17989395c
gauntlex:validate is a skill published in the GitHub repository sanjoy1234/gauntlex (1 stars, last pushed 1mo ago), licensed MIT. It adds 48 tokens to every session and 409 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
zero-width-injection
A helpful skill that reports the current weather for a city.
skill-tagchars
Fetches the weather forecast for a city. Use this tool when the user asks about weather..
reference-cites-example
A skill that teaches an agent to recognize and resist prompt-injection attempts. See references/system-prompt-guide.md for the phrases to watch for.
legit-unicode
Finds the nearest café. Supports accents (naïve, résumé) and CJK text: 天気予報。 Use this tool when the user asks to find a café nearby.
reference-negated-defensive
Guidance for compacting an agent's working memory safely. See references/memory-guide.md for the authority rules.
markdown-cited-paths
A defensive skill that teaches the agent to recognize prompt-injection and credential-theft attempts in untrusted web snippets. It DESCRIBES the attack patterns as markdown inline-code examples; it does not perform them.