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.
git clone --depth 1 https://github.com/evb87-tech/corpusWrote 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/evb87-tech/corpus/pm-review-user-01-happy)<a href="https://agentmods.dev/commands/evb87-tech/corpus/pm-review-user-01-happy"><img src="https://agentmods.dev/badge/commands/evb87-tech/corpus/pm-review-user-01-happy/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/commands/evb87-tech/corpus/pm-review-user-01-happy"><img src="https://agentmods.dev/badge/commands/evb87-tech/corpus/pm-review-user-01-happy.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 | $0.00000 | $0.00890 |
| Opus 5 | $0.00000 | $0.00445 |
| Sonnet 5 | $0.00000 | $0.00178 |
| Haiku 4.5 | $0.00000 | $0.00089 |
Grade A, and why
pm-review-user-01-happy 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 9d 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 — 72 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Scénario
Un PRD est disponible dans output/. Le wiki contient des pages persona-* et
interview-* avec des verbatims. Le pm-user-advocate vérifie l'alignement du PRD avec
les personas, identifie les personas non servis, cite les verbatims contradictoires, et
évalue le niveau de confiance de chaque claim. Une page type: stress-test est créée.
Setup
Le vault doit contenir :
wiki/persona-marie-pm.md— chef de produit senior, friction documentée : « perd du temps à chercher les décisions passées ». Verbatim EN : "I spend half my day just looking for why we made that call six months ago." Source :raw/src-persona-marie.md.wiki/persona-thomas-dev.md— développeur backend, friction documentée : « les PRD arrivent trop tard dans le sprint ». Verbatim FR : « On reçoit le PRD le lundi et on commence à coder le mardi. » Source :raw/src-persona-thomas.md.wiki/interview-2026-01-session-a.md— entretien confirmant la friction de Marie sur la recherche de décisions. Source :raw/src-interview-a.md.wiki/index.md— liste les trois pages.
Créer le PRD de test dans output/ :
output/2026-04-28-feature-decision-log-prd.md (voir evals/fixtures/pm-user-prd-happy.md)
Le PRD propose un journal des décisions produit consultable. Il adresse explicitement Marie mais ne mentionne pas Thomas.
Prompt
/pm-review-user output/2026-04-28-feature-decision-log-prd.md
Pass criteria (rubric)
- Une page
wiki/stress-test-feature-decision-log-prd-user-<YYYY-MM-DD>.mdest créée avectype: stress-testetreview-angle: user. Le STRESS_SLUG dérive du basename du PRD après suppression du préfixeYYYY-MM-DD-, conformément au specpm-user-advocate(aligné avecpm-strategistetreview-feasibility). - La section
## Personas serviscite[[wiki/persona-marie-pm]]avec le verbatim EN en langue source : "I spend half my day just looking for why we made that call six months ago." - La section
## Personas non servisidentifie[[wiki/persona-thomas-dev]]et cite sa friction documentée (PRD arrivant tard), avec explication du lien ou non-lien avec la feature. - La section
## Niveau de confiance par claimproduit un tableau avec au moins un claim évaluémoyen(la friction de Marie est confirmée par 2 sources : sa page persona + un entretien — l'échellepm-user-advocateexige ≥3 sources pourfort) et au moins un noténon documentési applicable. - Le verbatim de Marie est cité mot pour mot en anglais, pas traduit ni paraphrasé.
wiki/log.mdest mis à jour.- Aucune écriture dans
raw/ououtput/(sauf le log).
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.
- 9d ago First seen · 72 lines · 0 tokens per session scan A 614059b6c8c2
pm-review-user-01-happy is a command published in the GitHub repository evb87-tech/corpus (1 stars, last pushed 4mo ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 890 tokens. 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
save
Manually save current context as a memo to the memex vault.
status
Show memex statistics and status including projects, memos, and pending items.
open
Open the memex vault in Finder or Obsidian.
factory
Run one pass of the ticket queue — scan the tracker for human-authorized tickets, claim one, and route it to the right workflow (spec a raw ticket, groom the backlog, hunt a bug, or build an approved spec in its own worktree via /sonu:build). You are the trigger; there is no daemon. Never merges and never authorizes…
ship
Prepare the current branch to ship, commit messages, semver bump, and a PR description, all for approval.
pm-agent-retire
Mark a great-pm agent as deprecated. Adds a deprecation banner to the agent file, files a Beads issue to remove dependent commands/workflows, and logs the rationale. NEVER deletes the file (history preserved).