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 evanmcgoogan/skin-in-the-game --skill grill-megit clone --depth 1 https://github.com/evanmcgoogan/skin-in-the-gameWrote 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/evanmcgoogan/skin-in-the-game/grill-me)<a href="https://agentmods.dev/skills/evanmcgoogan/skin-in-the-game/grill-me"><img src="https://agentmods.dev/badge/skills/evanmcgoogan/skin-in-the-game/grill-me/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/evanmcgoogan/skin-in-the-game/grill-me"><img src="https://agentmods.dev/badge/skills/evanmcgoogan/skin-in-the-game/grill-me.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 | $0.00135 | $0.00700 |
| Opus 5 | $0.00068 | $0.00350 |
| Sonnet 5 | $0.00027 | $0.00140 |
| Haiku 4.5 | $0.00014 | $0.00070 |
Grade A, and why
grill-me 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.
How it starts
The opening of the file, as written. The whole thing — 34 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Grill Me
The most expensive failure in agent-assisted building is confidently building the wrong thing — filling ambiguity with your own assumptions instead of the human's intent. Grilling is the antidote, but it's a scalpel, not a hammer: it fires only at the CEO line — taste, strategy, direction, one-way doors. Below that line, asking is a breach of the contract; decide and move. (See operating-contract.)
When to grill
Grill when a substantial build hinges on a decision that is genuinely the human's:
- Taste — product feel, voice, identity
- Strategy — what to build, what to cut, positioning, priorities
- One-way doors — irreversible or expensive-to-reverse choices
Do NOT grill on architecture, library choices, naming, or anything reversible. If you can change it cheaply later, just decide.
How to grill well
The questions carry the value, so craft them from first principles:
- The options carry the insight. A weak question is "How should I handle errors?" A first-principles one is: "Failures here are either (a) recoverable → retry, (b) fatal → halt loudly, or (c) silent → log-and-continue. Which is acceptable?" The second forces a decision the human would otherwise discover at 2am.
- Aim at the one-way doors. Prioritize questions whose wrong answer is expensive to reverse — data model, public interface, what not to build.
- Each option states pros, cons, and long-term implications in plain English, and you lead with your recommendation and why.
- 3–5 questions, then build. It's not an interview; it's removing the top sources of rework.
Use the AskUserQuestion tool to present these as multiple choice — low friction for the human, and it forces you to enumerate the real, mutually-exclusive paths rather than asking an open "what do you want?"
Pair every grill with ownership
Grilling must never read as punting. Whenever you raise CEO-line questions, say in the same breath what you're already owning and proceeding on: "I'm owning the schema, the library, and the layout — I only need you on these two." A grill without that boundary is a runaway asking permission for everything; a grill with it is a COO bringing you the one decision that's actually yours. Decide the reversible, escalate the irreversible, and make both halves visible.
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 · 34 lines · 135 tokens per session scan A 06a5b529140a
grill-me is a skill published in the GitHub repository evanmcgoogan/skin-in-the-game (1 stars, last pushed 3mo ago), licensed MIT. It adds 135 tokens to every session and 700 once invoked, about $0.0007 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 2026-08-31.
Other skills, from other repositories
delegate
Let Claude Architect route a versioned implementation spec through the trusted MCP runtime, independently review the Candidate Artifact, record a decision, and integrate only accepted bytes. Use for implementation delegation, Producer selection, or commitment-boundary review.
codex
Run the Codex CLI directly in the user's checkout for code analysis, refactoring, or automated editing without Claude Architect's verified delegation lifecycle.
commit
A guided process for committing code changes to Git. It checks the working tree, follows a conventional commit format, links a task when available, reviews the changes, and asks for approval.
orient
Internal skill for commands. Render the shared orientation snapshot - active task, what's in flight, what's next - tracker-universal. Do not trigger on user conversation; only when a command needs the snapshot.
x-multi-llm-align
A review process in which two sub-agents examine an API, data format, event schema, or workflow from their separate implementation perspectives. The user passes documents and feedback between them over multiple rounds.
design
Create a doc-as-code design package from a PRD or SPEC. Conditionally generates C4 diagrams (Context/Container/Component), sequence diagrams, ER diagram + Data Dictionary, OpenAPI 3.0, AsyncAPI 3.0, ADRs, domain glossary, state diagrams, and deployment view as Mermaid-rendered Markdown files. Use when PM mentions…