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/punt-labs/z-spec/helpgit clone --depth 1 https://github.com/punt-labs/z-specWrote 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/punt-labs/z-spec/help)<a href="https://agentmods.dev/commands/punt-labs/z-spec/help"><img src="https://agentmods.dev/badge/commands/punt-labs/z-spec/help.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 | $0.00010 | $0.01750 |
| Opus 5 | $0.00005 | $0.00875 |
| Sonnet 5 | $0.00002 | $0.00350 |
| Haiku 4.5 | $0.00001 | $0.00175 |
Grade A, and why
help 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 4d 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 — 149 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Z Specification Plugin Help
Run this as an interactive getting-started flow. Do not display the material below as a static document — the goal is for the user to leave this conversation holding a fuzz-clean, model-checked Z specification of a problem they actually have, not a page they scrolled past.
Getting Started: Model Your Own Problem
Walk the user through these steps in order. Do not skip ahead to the command table below until this path is complete or the user asks for it directly.
If /z-spec:check or /z-spec:test refuses because z-spec isn't enabled
in this repository, run /z-spec:enable first, then resume the step that
was refused.
1. Find the stateful system
If the user's request already names a system to model, use it. Otherwise
ask: "What's a stateful system you'd like to model? Something with data
that changes over time and rules about what states are valid — a job
queue, a login session, a library's lending records, an inventory, a
workflow with stages." One or two sentences from the user is enough to
start; /z-spec:code2model will ask follow-up questions if the
description is ambiguous.
2. Draft the specification
Run /z-spec:code2model <their description>. This writes a first .tex
spec to docs/ with given sets, a state schema, an Init schema, and a
few operations — the raw material, not a finished spec.
3. Type-check it
Run /z-spec:check <the file>. fuzz will very likely report errors on a
first draft — that is normal, not a failure. Fix each one and re-run until
/z-spec:check reports OK. This loop, not the first draft, is where the
spec actually gets built.
4. Model-check it
Run /z-spec:test <the file>. probcli animates the spec and explores its
reachable states. If it reports a counter-example, that is the model
telling you something true about the specification as written — walk the
user through what the counter-example means and fix the spec (an
invariant, a precondition, a bound) rather than the report.
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.
- 4d ago First seen · 149 lines · 10 tokens per session scan A e7216f01ec55
help is a command published in the GitHub repository punt-labs/z-spec (5 stars, last pushed today), licensed MIT. It adds 10 tokens to every session and 1,750 once invoked, about $0.0001 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
meeting-listen
Play back a completed meeting summary as a voiced debate between personas.
meeting-hive
Run an autonomous PR/FAQ review meeting where four personas debate and reach consensus without user intervention.
vote
Assess whether a PR/FAQ should move forward with a structured go/no-go decision.
feedback
Incorporate feedback into PR/FAQ and redraft affected sections.
feedback-to-us
Tell us how the prfaq plugin is working for you (anonymous 1-5 feedback).
badge
Generate a stage-colored badge and embed it in your README.