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 Rebel028/gauntlet --skill skillgit clone --depth 1 https://github.com/Rebel028/gauntletWrote 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/rebel028/gauntlet/skill)<a href="https://agentmods.dev/skills/rebel028/gauntlet/skill"><img src="https://agentmods.dev/badge/skills/rebel028/gauntlet/skill/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/rebel028/gauntlet/skill"><img src="https://agentmods.dev/badge/skills/rebel028/gauntlet/skill.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 | $0.00000 | $0.00778 |
| Opus 5.5 | $0.00000 | $0.00311 |
| Sonnet 5.5 | $0.00000 | $0.00156 |
| Haiku 4.5 | $0.00000 | $0.00078 |
Grade A, and why
skill 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 1mo 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 — 37 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Gauntlet
A bias-check technique: spawn 2+ independent subagents, each attacking the idea from a different angle, to catch the flaw BEFORE it's set in concrete.
The name is the method: you run the idea through a gauntlet of independent attackers, each trying to stop it for a different reason. What makes it out the far side has earned your commitment. When you're choosing between options, run them all through — the one that takes the most hits and survives is usually the answer.
When picking between options rather than vetting one, run each agent against all the options at once ("attack each of these three caching approaches; for each, find where it breaks") so the comparison stays apples-to-apples.
How to run it
-
Pick 2+ agents with DIFFERENT angles. Not "one for / one against" — that's a weak debate. You want genuinely different optics so the attacks don't overlap. The five standing personas are all read-only (they ground critique in your code but never modify it):
gauntlet-formalist— structure, logic, invariants. Are the categories actually orthogonal?gauntlet-practitioner— the year-one maintainer. What gets painful, who gets paged?gauntlet-consumer-advocate— the downstream consumer (API client, on-call). What breaks for them?gauntlet-historian— prior art. Who tried this and why did it break?gauntlet-threat-modeler— security. Where's the abuse case, the blast radius?
No fit? Use the
custom-adversarytemplate (custom-adversary.md, alongside this skill) — not a registered agent; paste it into a general-purpose subagent restricted to read-only tools.Pick angles that fit the decision. A migration → formalist + practitioner + historian. An auth change → threat-modeler + consumer-advocate + formalist. Pick the 2–4 sharpest, not all six.
-
Give each agent the FULL idea: what it is, what's decided, which alternatives were rejected and why, what's irreversible. Without it the attack misses.
-
Do NOT hint at the conclusion you want. Never write "confirm that X is good." Write "attack X, find where it breaks." A skeptic agent that's been told the desired answer is useless.
What ships with it
1 file 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.
- 1mo ago Changed 58e75aa53bb2
- 2mo ago First seen · 37 lines · 0 tokens per session scan A c416e8081a05
skill is a skill published in the GitHub repository Rebel028/gauntlet (6 stars, last pushed 1mo ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 778 tokens. 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
subagent-driven-development
Execute plans via delegatetask subagents (2-stage review).
duckduckgo-search
Free keyless web, news, and image search via ddgs.
mcporter
List, auth, and call MCP servers/tools from the terminal.
article-writing
Write articles, guides, blog posts, tutorials, newsletter issues, and other long-form content in a distinctive voice derived from supplied examples or brand guidance. Use when the user wants polished written content longer than a paragraph, especially when voice consistency, structure, and credibility matter.
mem0-oss-to-platform
Plan, then execute, a migration of a project from the mem0 open-source / self-hosted SDK (local Memory class) to the mem0 Platform / hosted SDK (MemoryClient). TRIGGER when: the developer wants to move, switch, or migrate mem0 off OSS/self-hosted to the hosted API, e.g. "migrate my mem0 setup to the platform", "switch…
safe-refactor
Restructure code while preserving behavior. Use for extraction, consolidation, ownership moves, or cleanup where verification must bracket structural edits.