draft-01-happy

A test scenario for creating a presentation brief from a research wiki. The brief must describe a target user, a product-discovery method, and the expected deliverables.

In plain words
What is it for?
Testing a brief for a senior engineering team using pages about a product manager persona, jobs to be done, and product discovery.
Why use it?
It checks that the draft is saved as an output, cites the relevant wiki pages, and contains only claims supported by those pages.

Command

Install

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.

agentmods
npx agentmods add commands/evb87-tech/corpus/draft-01-happy
Clone the repo
git clone --depth 1 https://github.com/evb87-tech/corpus
Per session 0 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 529 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5 $0.00000 $0.00529
Opus 5 $0.00000 $0.00264
Sonnet 5 $0.00000 $0.00106
Haiku 4.5 $0.00000 $0.00053

Measured 2d ago against content hash faa8e7b4cf78, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

draft-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 2d 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.

corpus-core/evals/commands/draft-01-happy.md · 54 lines

What it actually says

Scénario

Le wiki contient suffisamment de matière pour rédiger un brief de présentation. L'utilisateur demande un draft. Le comportement attendu est qu'un fichier est créé dans output/, citant les pages wiki pertinentes, sans écrire dans wiki/ et sans compléter avec des connaissances d'entraînement.

Setup

Le vault doit contenir :

  • wiki/persona-marie-pm.md — fiche persona d'une chef de produit senior. Source : raw/src-persona-marie.md.
  • wiki/jobs-to-be-done.md — description du cadre JTBD. Source : raw/src-jtbd.md.
  • wiki/discovery-produit.md — phases de la discovery. Source : raw/src-discovery.md.
  • wiki/index.md — liste les trois pages.

Prompt

/draft "Brief de présentation d'une démarche de discovery produit à destination d'une
équipe engineering senior. Inclure : le persona ciblé, la méthodologie, et les livrables
attendus."

Pass criteria (rubric)

  • Un fichier est créé dans output/ (ex. output/YYYY-MM-DD-brief-discovery-engineering.md).
  • Le fichier cite les pages wiki avec [[wiki/persona-marie-pm]], [[wiki/jobs-to-be-done]], [[wiki/discovery-produit]].
  • Toutes les affirmations du brief sont traçables aux pages wiki citées.
  • Aucune page n'est créée ni modifiée dans wiki/.
  • Le récapitulatif final liste le fichier output produit, les pages wiki citées, et les éventuelles lacunes signalées.
  • Si une information nécessaire au brief n'est pas dans le wiki, le modèle s'arrête et demande confirmation avant de compléter (il ne comble pas silencieusement).

Fail signals

  • Une page est créée dans wiki/ (ex. wiki/brief-discovery.md) — violation de la règle « output/ ne revient pas dans wiki/ ».
  • Une affirmation du brief (ex. une statistique sur la discovery, un exemple de deliverable) provient de connaissances d'entraînement et n'est pas présente dans les pages wiki.
  • Le fichier output ne contient aucune référence [[wiki/...]].
Changes

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.

  1. 2d ago First seen · 54 lines · 0 tokens per session scan A faa8e7b4cf78

Subscribe to this mod's changes

draft-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 529 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.