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.
git clone --depth 1 https://github.com/emaraschio/cursor-commandsWrote 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/commands/emaraschio/cursor-commands/gauntlet-loop)<a href="https://agentmods.dev/commands/emaraschio/cursor-commands/gauntlet-loop"><img src="https://agentmods.dev/badge/commands/emaraschio/cursor-commands/gauntlet-loop/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/commands/emaraschio/cursor-commands/gauntlet-loop"><img src="https://agentmods.dev/badge/commands/emaraschio/cursor-commands/gauntlet-loop.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 | $0.00024 | $0.01667 |
| Opus 5.5 | $0.00010 | $0.00667 |
| Sonnet 5.5 | $0.00005 | $0.00333 |
| Haiku 4.5 | $0.00002 | $0.00167 |
Grade A, and why
gauntlet-loop 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 — 66 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Overview
Run a Gauntlet Loop: take a GOAL and a REAL-WORLD EQUIVALENT with an inspectable reference pack, decompose into independent parts, and run each part through a build → critique → pass|iterate|terminal loop. After all parts pass, run an integration critic on the whole. Optional BUDGET ceiling, taste-domain pride gate, and gap ledger for resume. Critics use fresh context, inspect the artifact, and compare against the pack (blind when possible). A part passes only when better than the reference (equal fails); otherwise the critic returns the largest specific gap. Builders never evaluate their own work. Full workflow: .cursor/skill-contracts/gauntlet-loop/SKILL.md (user install: ~/.cursor/skill-contracts/gauntlet-loop/SKILL.md).
Defaults
| Setting | Default |
|---|---|
| Intake | GOAL, REAL-WORLD EQUIVALENT, and inspectable reference pack required |
| Iteration cap | 5 rounds per part (overridable at intake) |
| BUDGET | None beyond per-part cap (optional total rounds / wall-clock) |
| Stall | Same gap twice with no improvement; halt that part |
| Critic mode | Subagent when available; else fresh-context simulation (report which) |
| Blind comparison | Prefer side-by-side without labeling which is the reference |
| Integration critic | Required after all parts pass before done |
| Pride gate | Taste domains only (UI, game feel, writing); else "not a taste domain" |
| Gap ledger | Record round/gap/verdict; consult on resume |
| Part list | Publish then proceed unless the user objects |
| Examples | Generic names only (project-a, game-1) |
Steps
- Read
.cursor/skill-contracts/gauntlet-loop/SKILL.mdfor the full agent contract; if that file is missing, read~/.cursor/skill-contracts/gauntlet-loop/SKILL.md. - Execute phases in order (Intake → Decompose → Per-part loop → Integration critic → Deliver).
- Report the part table, integration verdict, gap ledger, pride gate (or not a taste domain), and halt on stalled, capped, or budgeted parts; do not claim done from part-local passes alone.
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 First seen · 66 lines · 24 tokens per session scan A 170c32ab6b7f
gauntlet-loop is a command published in the GitHub repository emaraschio/cursor-commands (9 stars, last pushed 19d ago), licensed MIT. It adds 24 tokens to every session and 1,667 once invoked, about $0.0001 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 commands, from other repositories
green-repo
Drive the whole repository to a verified-green baseline — typecheck, lint, tests, build — with authorized fixing of pre-existing debt.
test
Route to the right test skill: unit, headed Playwright, QA, exploratory, visual regression, load, or mobile emulator.
arch-boundaries
Install mechanically-enforced architecture boundaries that fail CI on spaghetti imports.
test-plan
User-story-driven test coverage audit + plan — no tests written until approved.
test-mutation
Set up mutation testing to prove tests assert behavior, not just coverage.
check-coverage
Check test coverage, identify gaps, and suggest which code needs tests.