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 agentmods add commands/runnithan/claudex-setup/quiz-megit clone --depth 1 https://github.com/runnithan/claudex-setupWrote 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/commands/runnithan/claudex-setup/quiz-me)<a href="https://agentmods.dev/commands/runnithan/claudex-setup/quiz-me"><img src="https://agentmods.dev/badge/commands/runnithan/claudex-setup/quiz-me.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 | $0.00046 | $0.01014 |
| Opus 5 | $0.00023 | $0.00507 |
| Sonnet 5 | $0.00009 | $0.00203 |
| Haiku 4.5 | $0.00005 | $0.00101 |
Grade A, and why
quiz-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 5d 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 — 80 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/quiz-me: examine me on this change
You are reviewing by examination, not by summary. Reading an agent-written diff line by line does not scale with the rate agents produce code, and a summary tells me what the author thinks happened. A quiz inverts it: the gaps in my understanding surface actively, and the questions I cannot answer are the review findings. You are strictly read-only: never edit, fix, commit, or run anything that changes state.
Arguments: $ARGUMENTS
1. Resolve the change
Interpret the arguments in this order:
- a git range (
main..HEAD,abc123..def456) or a single commit/ref: usegit diff/git show; - a PR number or URL:
gh pr diff <n>for the diff andgh pr view <n> --json title,body,filesfor the framing; - nothing: the working tree diff (
git diffplusgit diff --cached); if that is empty, the last commit (git show HEAD). --n Nsets the number of questions (default 6).--diagram-onlystops after step 3.
State what you resolved in one line ("Quizzing on main..HEAD, 14 files, +612/-88") before going on.
2. Read it properly
Read the whole diff, then enough of the surrounding code to know: which callers are affected, which invariants or data shapes changed, where the error paths go, what the changed tests actually prove, and what the diff looks like it touches but does not. Do not quiz from the diff alone; a question about a caller you never opened is a guess.
3. Orient me
One paragraph: what moved and why, in plain language. Then a compact diagram (mermaid, or ASCII if mermaid would not render here) of the touched components and the edges that are new or changed. Mark removed edges too. Keep it to what the diff changed; this is a map of the change, not of the system.
4. Quiz
Ask N questions, all at once, numbered, ordered from structural to edge cases. Every question
must be one a reviewer has to be able to answer before approving, and prefer questions whose
wrong answer would be a merge mistake:
- why this was changed, not just what;
- what happens when a specific input, state, or failure occurs on the new path;
- which callers or consumers are affected, and which one is easy to miss;
- what a changed or added test actually proves, and what it does not;
- what was deliberately left alone, and why that is safe;
- plus one trap: something the diff looks like it does but does not (or the reverse).
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.
- 5d ago First seen · 80 lines · 46 tokens per session scan A b4504288f3a1
quiz-me is a command published in the GitHub repository runnithan/claudex-setup (1 stars, last pushed 4d ago), licensed MIT. It adds 46 tokens to every session and 1,014 once invoked, about $0.0002 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 commands, from other repositories
factcheck
Verify BibTeX entries and cited claims against DBLP and web sources. Checks author names, venues, years, and specific numerical claims.
translate-review-to-single-human
Translate a multi-reviewer code review into a single human-voice GitHub PR comment.
address
Address code review feedback — corroborate, validate, and implement changes from a review's final.md.
map
Generate a Code Review Map to help navigate large, complex changesets.
post
Post the current OCR review to a GitHub PR.
show
Display a past OCR review session.