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 paulbaranowski/wild-horses --skill steelmangit clone --depth 1 https://github.com/paulbaranowski/wild-horsesWrote 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/paulbaranowski/wild-horses/steelman)<a href="https://agentmods.dev/skills/paulbaranowski/wild-horses/steelman"><img src="https://agentmods.dev/badge/skills/paulbaranowski/wild-horses/steelman/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/paulbaranowski/wild-horses/steelman"><img src="https://agentmods.dev/badge/skills/paulbaranowski/wild-horses/steelman.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 | $0.00090 | $0.00950 |
| Opus 5 | $0.00045 | $0.00475 |
| Sonnet 5 | $0.00018 | $0.00190 |
| Haiku 4.5 | $0.00009 | $0.00095 |
Grade A, and why
steelman 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 10d 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 — 54 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Steelman the Case Against
Your job is to argue against the proposed changes — and to make that argument as strong as an intelligent, well-informed opponent would. This is a steelman, not a strawman: you are constructing the objection the proponent should most fear, not scoring cheap points.
You are a temporary adversary on purpose. Do not hedge into balance, do not end with "but it's probably fine." Someone else owns the case for the change; for this turn, you own the case against it.
Step 1 — Identify what you're arguing against
- If
$ARGUMENTSnames a file or path — read it. That document is the proposal. - If
$ARGUMENTSis empty (DEFAULT) — the proposal is whatever change is currently on the table in this conversation: the plan, design, spec, or diff most recently discussed. If that's genuinely ambiguous, ask one yes/no question to pin it down before continuing.
Restate the proposal in one or two sentences so it's clear what you're attacking. If you can't restate it faithfully, you can't steelman against it — get the restatement right first.
Step 2 — Build the strongest case against
Work through these angles. Use the ones that bite; skip the ones that don't. Quality over coverage — three sharp objections beat ten weak ones.
- Hidden / underestimated cost — what this actually costs to build, maintain, document, and operate that the proposal glosses over.
- Load-bearing assumptions — the premises the change depends on. Name each one and ask what happens if it's wrong.
- Simpler alternative — a cheaper way to get most of the benefit, including changing nothing. If a smaller move captures 80% of the value, the full change has to justify the remaining 20%.
- Second-order effects — what this makes harder later: new coupling, precedent, migration debt, things future changes now have to route around.
- Reversibility — how hard this is to undo once shipped. One-way doors deserve more suspicion than two-way doors.
- Who pays — costs that land on someone other than the proposer (other maintainers, users, on-call, future-you).
- The do-nothing option — what's actually wrong with the status quo, honestly weighed. Sometimes the current state is fine and the itch isn't worth scratching.
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.
- 10d ago First seen · 54 lines · 90 tokens per session scan A 245948db5567
steelman is a skill published in the GitHub repository paulbaranowski/wild-horses (12 stars, last pushed 22d ago), licensed MIT. It adds 90 tokens to every session and 950 once invoked, about $0.0005 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 2026-08-30.
Other skills, from other repositories
autoreview
Pre-commit/ship code review: Codex default; optional Claude or Pi.
rework-rate
Measure and interpret PR rework rate — the emerging 5th DORA metric.
omh-code-review
This is a Hermes-native code-review workflow skill.
revdiff-plan
Review the last Codex assistant message (plan, analysis, or proposal) with inline annotations in a TUI overlay. Extracts the most recent response from Codex rollout files and opens it in revdiff for review and annotation. Activates on "revdiff-plan", "review plan with revdiff", "annotate plan", "review last response"…
code-reviewer
Code review specialist focused on patterns, bugs, security, and performance.
full-repo-review
Comprehensive four-wave review of all repo source files, producing a prioritized issue backlog.