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/Jott2121/agent-gateWrote 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/rules/jott2121/agent-gate/30-benchmarks-and-claims)<a href="https://agentmods.dev/rules/jott2121/agent-gate/30-benchmarks-and-claims"><img src="https://agentmods.dev/badge/rules/jott2121/agent-gate/30-benchmarks-and-claims/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/rules/jott2121/agent-gate/30-benchmarks-and-claims"><img src="https://agentmods.dev/badge/rules/jott2121/agent-gate/30-benchmarks-and-claims.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 | $0.00323 |
| Opus 5.5 | $0.00000 | $0.00129 |
| Sonnet 5.5 | $0.00000 | $0.00065 |
| Haiku 4.5 | $0.00000 | $0.00032 |
Grade A, and why
30-benchmarks-and-claims 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
Benchmarks and public claims
- Pre-register before comparing: hypothesis, metric, n, decision rule →
docs/PREREG.md, committed before the run. - Every published number links to: the command, the commit SHA, the raw artifact, and the date.
- Include the baseline and the null. If a result is not significant, say so in the headline, not a footnote.
- State limits up front (n, single run per cell, prompt shape, model versions, cost).
- Don't round in your favor. Show CIs or ranges where you have them.
- Never edit raw run outputs. Supersede with a new run and a note.
- If a README number changes, the PR must show before → after and the command that produced it.
- Cross-check numbers that appear in more than one place (repo README, profile README, LinkedIn, posts) and fix drift in the same PR.
- This repo has no
eval/harness. The local measured claim is the coverage sentence in README.md. The16.7%figure is a citation of sabot, not a number this package recomputes. Demo hash prefixes are not stable receipts: ledger hashes include a wall-clock timestamp.
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 · 17 lines · 0 tokens per session scan A d64078af151e
30-benchmarks-and-claims is a cursor rule published in the GitHub repository Jott2121/agent-gate (3 stars, last pushed yesterday), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 323 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 2026-10-12.
Other cursor rules, from other repositories
ponytail
Ponytail, lazy senior dev mode. Always pick the simplest solution that works.
angular-20
This rule provides comprehensive best practices and coding standards for Angular development, focusing on modern TypeScript, standalone components, signals, and performance optimizations.
dev-standard
Apache Superset development standards and guidelines for Cursor IDE.
prefer-assertions-over-defensive-checks
Prefer assertions over defensive checks when data is guaranteed to be valid.
prefer-direct-imports-over-module-mocks
Prefer extracting a testable core over vi.mock / vi.resetModules when unit tests need to reach production logic entangled with config, env, or singletons.
shared-plane-packages
Pattern for creating shared plane packages that serve both migration and runtime planes.