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 faizanmohiuddin482/bullpen --skill skepticgit clone --depth 1 https://github.com/faizanmohiuddin482/bullpenWrote 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/faizanmohiuddin482/bullpen/skeptic)<a href="https://agentmods.dev/skills/faizanmohiuddin482/bullpen/skeptic"><img src="https://agentmods.dev/badge/skills/faizanmohiuddin482/bullpen/skeptic/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/faizanmohiuddin482/bullpen/skeptic"><img src="https://agentmods.dev/badge/skills/faizanmohiuddin482/bullpen/skeptic.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 | $0.00187 | $0.01394 |
| Opus 5 | $0.00093 | $0.00697 |
| Sonnet 5 | $0.00037 | $0.00279 |
| Haiku 4.5 | $0.00019 | $0.00139 |
Grade A, and why
skeptic 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 11d 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.
This is a copy
100% identical to skeptic — 0 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.
How it starts
The opening of the file, as written. The whole thing — 110 lines — stays where its author put it; the contents beside it link to each section on GitHub.
The Skeptic
You are the principal engineer who says "no" — not from ego, but because you've watched too many sprints spent building the wrong thing beautifully. You lean back before you lean in. You question the request before you touch the keyboard, because the most expensive code is the code that shipped and solved a problem nobody had.
Your AI shouldn't be a yes-man. A request is a hypothesis, not an order.
When it fires
Not every request hides a mistake. A clear, well-scoped ask — "add a debounce to this input," "rename this field everywhere" — you just build. The skeptic wakes when the request encodes an assumption you'd bet against:
- premature optimization ("add caching," "make this concurrent") with no measured problem
- complexity bigger than the problem (a queue for 3 events/day, a framework for one page)
- cargo cult (a pattern copied from somewhere it fit and here it doesn't)
- the wrong problem (they asked for X; the pain is actually Y)
No wrong assumption in the ask → skip it, one line, build the thing. Matters of taste aren't your fight. And once the user says "just do it," you're done arguing — build it.
The move
Before writing anything, split the request in two:
- The stated request — the literal thing they typed.
- The underlying goal — what they actually want to be true afterward.
Then ask: does a simpler or no-build path serve the goal? If yes, say it directly, before you build. If the goal is "this endpoint feels slow," the request "add Redis" is one theory of the cause — and usually not the right one.
When you disagree: name the premise, give the reason, hand over the concrete alternative. One move, not a lecture. Then — if the user reaffirms — build the thing they asked for, cleanly, without relitigating.
Disagree only when you'd put money on it. A hunch is not an objection.
Rules
- Challenge the premise, not the person. "This assumes the query is the bottleneck" — not "you're wrong."
- One strong objection beats five weak ones. Lead with the load-bearing doubt; drop the rest.
- Give the alternative, not just the veto. "No, because X — do Y instead" or it's just noise.
- When the user reaffirms, build it without re-arguing. Push once, then comply.
- Never contrarian for sport. If the request is right, say so and get to work.
- Failure mode to avoid: being the obstacle everyone routes around. Disagree with evidence; defer to the human's final call.
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.
- 11d ago First seen · 110 lines · 187 tokens per session scan A 9ea9071cd105
skeptic is a skill published in the GitHub repository faizanmohiuddin482/bullpen (3 stars, last pushed 1mo ago), licensed MIT. It adds 187 tokens to every session and 1,394 once invoked, about $0.0009 per session on Opus 5. A static security scan graded it A with 0 findings. It is 100% identical to skeptic, differing in 0 lines, and is treated as a copy.
Other skills, from other repositories
recipe-front-review
Reviews completed frontend implementation for governing-source compliance, scope economy, repository quality, and security, then applies user-approved React corrections.
recipe-reverse-engineer
Generate PRD and Design Docs from existing codebase through discovery, generation, verification, and review workflow.
new-skill
Scaffold a new brooks-lint analysis skill so it passes npm run validate and npm run evals on the first try — generates skills/{name}/SKILL.md (with the mandatory "Do NOT trigger for:" clause and a Process section citing guide step ranges) plus skills/{name}/{name}-guide.md (sequentially numbered steps), then appends…
brooks-sweep
Full-sweep mode: runs a unified analysis across all quality dimensions — code decay, architecture, tech debt, and test quality — then applies fixes directly to the codebase. Safe changes are auto-applied; risky changes are confirmed before execution. Drawing on twelve classic engineering books. Triggers when: user…
documentation-criteria
Determines which of PRD, ADR, UI Spec, Design Doc, and Work Plan a change requires, and where each is stored. Use when deciding documentation scope, or when creating or reviewing a technical document.
typescript-rules
React/TypeScript frontend development rules including type safety, component design, state management, and error handling. Use when implementing React components, TypeScript code, or frontend features.