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/steve-magne/hookstack/implementgit clone --depth 1 https://github.com/steve-magne/hookstackWhat 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 | $0.00000 | $0.01413 |
| Opus 5 | $0.00000 | $0.00707 |
| Sonnet 5 | $0.00000 | $0.00283 |
| Haiku 4.5 | $0.00000 | $0.00141 |
Grade A, and why
implement 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 yesterday.
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 — 78 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Implement — Planner → Generator → Evaluator
Conçoit (si user-facing) puis livre une feature, selon le pattern à 3 rôles séparés (anti-biais d'auto-éval). Deux sources d'entrée — le mode est déduit de l'argument :
$ARGUMENTS =
- une idée libre → mode idée (flux rapide) :
/implement "ajouter un filtre par provider sur le catalogue". Raffinage live + dev immédiat dans cette session, sans passer par GitHub. - un numéro / URL d'issue → mode issue :
/implement 88. Lire son contenu avecgh issue view <N>pour cadrer le Contrat d'implémentation.
0. Cadrage du contexte projet (déterministe, ~0 token)
Avant de planifier, s'ancrer dans le projet sans charger de code en contexte principal :
node scripts/okf.mjs query <termes de la feature> (retrieval déterministe). N'escalader au
sous-agent okf-librarian (lit dans son propre contexte → renvoie une synthèse courte) que si la
query ne suffit pas. Objectif : le Planner connaît contraintes + patterns existants (Motion,
registre, CLI, conventions, cohérence 4 surfaces) avant d'écrire le Contrat — pas de lecture
exploratoire du code en contexte principal. C'est le levier principal d'économie de tokens :
chaque lecture lourde se fait dans le contexte d'un sous-agent, jamais dans le tien.
0.5 Design Brief (conception multi-perspectives) — features user-facing uniquement
Sauter pour un fix purement technique (pas d'écran/UX impacté) — dis-le en une ligne. Sinon, fan-out parallèle sur des lentilles indépendantes, chacune lit son OKF dans son propre contexte :
frontend-engineer→ où l'utilisateur le voit, hiérarchie visuelle, Motion (m.*, tokensmotion.ts), accessibilité.- lentille produit/cohérence (
general-purposeouokf-librarian) → sert la promesse produit ? Impact sur les 4 surfaces (site / CLI /doc/product//README.md) ?
Chaque lentille renvoie 2-3 lignes max (proposition + 1 risque). Tu synthétises un Design Brief de 5-8 lignes (où ça vit · feedback clé · 1 garde-fou UX ou de cohérence). Il devient un input du Contrat d'implémentation ci-dessous. Garde-fou anti-scope-creep : le brief sert l'intention de départ (l'idée, ou le périmètre de l'issue), il ne la gonfle pas.
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.
- yesterday First seen · 78 lines · 0 tokens per session scan A 1a9011dd6e75
implement is a command published in the GitHub repository steve-magne/hookstack (7 stars, last pushed 8d ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 1,413 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
checklist
Generate a custom checklist for the current feature based on user requirements.
clarify
Identify underspecified areas in the current feature spec by asking up to 5 highly targeted clarification questions and encoding answers back into the spec.
specify
Create or update the feature specification from a natural language feature description.
analyze
Perform a non-destructive cross-artifact consistency and quality analysis across spec.md, plan.md, and tasks.md after task generation.
converge
Assess the current codebase against the feature's spec, plan, and tasks, then append any remaining unbuilt work as new tasks to tasks.md so implement can complete it.
implement
Execute the implementation plan by processing and executing all tasks defined in tasks.md.