Getting it into your agent
It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.
git clone --depth 1 https://github.com/mthines/agent-skillsnpx agentmods add skills/mthines/agent-skills/eval-iterateWrote 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/mthines/agent-skills/eval-iterate)<a href="https://agentmods.dev/skills/mthines/agent-skills/eval-iterate"><img src="https://agentmods.dev/badge/skills/mthines/agent-skills/eval-iterate.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 | $0.00252 | $0.04409 |
| Opus 5 | $0.00126 | $0.02204 |
| Sonnet 5 | $0.00050 | $0.00882 |
| Haiku 4.5 | $0.00025 | $0.00441 |
Grade A, and why
eval-iterate 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 3d 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 — 357 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Eval Iterate
Drive a failing AI/LLM eval to a real green: diagnose, fix, re-run, confirm — capped at 5 iterations, never by weakening the eval.
This SKILL.md is the orchestration index.
Load the matching rule file when you need detail — do not preload them.
| Phase | Goal | Required rule |
|---|---|---|
| 0 | Resolve the target eval + capture the baseline failure | this file |
| 1 | Resolve how to run it | this file |
| 2 | Classify the failure (verdict required) | rules/eval-bug-classification.md |
| 3 | Apply the minimal fix — gated if it touches the eval itself | rules/anti-gaming-guard.md |
| 4 | Re-run, then confirm with a second run | rules/convergence-confirmation.md |
| 5 | Iterate or stop at the cap | this file |
| 6 | Report (structured exit summary) | this file |
Always read rules/anti-gaming-guard.md
before touching any eval definition (assertion, threshold, golden-set
item, judge prompt). The refusals in it apply on every iteration.
Input
The user provides one of:
- An eval identifier — an L2 suite name (e.g.
tier-routing), a golden-set file path, or a test/eval file path. - A PR URL with a failing eval check.
- Nothing — if
$ARGUMENTSis empty, auto-detect the failing eval check on the current branch's open PR (see Phase 0). --max-iterations <n>— lowers the cap below 5. Never raises it. A value above 5 is clamped to 5, not honored.
The argument is: $ARGUMENTS.
Phase 0 — Resolve the target + capture the baseline
If $ARGUMENTS is empty, do not ask the user — resolve automatically:
- Get the current branch and its open PR:
git rev-parse --abbrev-ref HEAD gh pr list --head "<branch>" --state open --json number,url --limit 1 - List failing checks and find the one that is an eval (name contains
eval,l2, or matches a known suite):gh pr checks <pr-number> --repo <owner/repo> - If exactly one failing eval check is found, use it as the target. If more than one, list them and ask the user which to iterate on first — this skill iterates on one target at a time. If none is found, report that and stop; there is nothing to iterate on.
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.
- 3d ago First seen · 357 lines · 252 tokens per session scan A 5d2632e91d32
eval-iterate is a skill published in the GitHub repository mthines/agent-skills (12 stars, last pushed yesterday), licensed MIT. It adds 252 tokens to every session and 4,409 once invoked, about $0.0013 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 2026-09-03.
Other skills, from other repositories
coverage-tracker
Run a Google Alerts-style keyword coverage tracker. Uses news-search for recent keyword queries, lets the LLM dedupe and classify real features versus junk, stores decisions in SQLite, and alerts only on new real coverage.
prompt-set-qa
Gate a prompt universe for schema and provenance completeness, target or campaign contamination, evidence entailment, naturalness, one-concept clarity, architecture consistency, aided status, answer leakage, and semantic duplicates. Use after realistic prompt generation and before human panel selection.
realistic-prompt-generation
Generate natural, controlled prompt variants from a target-blind design brief and prompt architecture while preserving approved jobs, acts, journeys, constraints, roles, locales, proximity bands, and evidence language. Use after architecture design and before contamination or semantic QA.
ai-visibility-panel-design
Select QA-approved canonical intent cells into a versioned AI-visibility tracking panel with partitions, variants, lanes, surfaces, locales, repetitions, separate exposure and priority weights, randomization, uncertainty, refresh rules, and campaign controls. Use after prompt QA or when revising an existing panel.
smoke-test-ml-pipeline
Owns the smoke test contract for an ML experiment: a small, diagnostic-by-construction pytest that fits the experiment's learner on a portion of the real data/ source and predicts on a disjoint portion that deliberately carries no pre-history buffer. The assertion is structural — the number of predictions must equal…
write-kaggle-benchmarks
Write, push, run, publish, and manage Kaggle Benchmark tasks using the kaggle CLI and the kaggle-benchmarks Python SDK. Use when the user wants to create or push a benchmark task (optionally with attached Kaggle datasets), run benchmarks against LLM models, check task/run status, stream or fetch execution logs…