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 skills add steve-magne/hookstack --skill tour-de-controlegit clone --depth 1 https://github.com/steve-magne/hookstackWrote 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/skills/steve-magne/hookstack/tour-de-controle)<a href="https://agentmods.dev/skills/steve-magne/hookstack/tour-de-controle"><img src="https://agentmods.dev/badge/skills/steve-magne/hookstack/tour-de-controle/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/skills/steve-magne/hookstack/tour-de-controle"><img src="https://agentmods.dev/badge/skills/steve-magne/hookstack/tour-de-controle.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00159 | $0.02254 |
| Opus 5.5 | $0.00064 | $0.00902 |
| Sonnet 5.5 | $0.00032 | $0.00451 |
| Haiku 4.5 | $0.00016 | $0.00225 |
Grade A, and why
tour-de-controle 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 1mo 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 — 131 lines — stays where its author put it; the contents beside it link to each section on GitHub.
tour-de-controle — planifier, déléguer, vérifier
Le principe : le grand modèle est trop précieux pour exécuter, et trop bon pour ne pas vérifier. Sur une grosse tâche, on ne lui demande pas de tout faire — on lui demande les deux choses où il écrase tout : le plan et le contrôle qualité. L'exécution, elle, part vers des sous-agents plus petits, plus rapides, moins chers, qui travaillent en parallèle.
PLANIFIER DÉLÉGUER VÉRIFIER
(grand modèle, 1 fois) (petits agents, en //) (grand modèle, 1 fois)
┌──────────────────┐ ┌──────┐ ┌──────┐ ┌──────────────────┐
│ spec complète │ → │lot 1 │ │lot 2 │ ... → │ contrôle contre │
│ + critères de │ └──────┘ └──────┘ │ les critères du │
│ "fini" par lot │ chaque lot = un agent │ plan, lot par lot│
└──────────────────┘ autonome └──────────────────┘
↳ ce qui échoue
repart en lot ciblé
Pourquoi ça marche (et pourquoi maintenant) : → Les modèles frontière type Claude Fable 5 sont devenus excellents en planification long-horizon et en vérification — mais chaque tour coûte cher et peut durer plusieurs minutes. Les utiliser pour exécuter chaque sous-tâche, c'est payer un prix de tour de contrôle pour faire rouler des chariots à bagages. → Un petit modèle avec une spec précise livre un travail quasi équivalent sur un lot bien découpé. Toute la qualité se joue dans la découpe et dans le contrôle — pas dans le muscle de l'exécutant. → La vérification par un agent à contexte frais bat l'auto-critique : l'exécutant ne voit pas ses propres angles morts.
Quand l'utiliser
- Tâche en plusieurs lots indépendants : audit d'un dossier, migration de fichiers, production d'une série de documents, analyse multi-sources, refonte par sections.
- L'utilisateur veut borner le coût tout en gardant la qualité du grand modèle là où elle compte.
- La tâche a des critères de réussite vérifiables (ou peut en recevoir).
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 1mo ago First seen · 131 lines · 159 tokens per session scan A 887b5b82c19e
tour-de-controle is a skill published in the GitHub repository steve-magne/hookstack (7 stars, last pushed 2d ago), licensed MIT. It adds 159 tokens to every session and 2,254 once invoked, about $0.0006 per session on Opus 5.5. 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 skills, from other repositories
{{BUCKET}}-management
Skill "{{BUCKET}}-management" from damusix/atomic-claude, covering purpose, doc shape and frontmatter contract.
vc:team
Orchestrate Agent Teams for parallel multi-session collaboration. Use for research, implementation, review, and debug workflows requiring independent teammates.
story
Track tickets, issues, and progress for your project. Load project context, manage sessions, guide setup.
story
Track tickets, issues, and progress for your project. Load project context, manage sessions, guide setup.
vc:watzup
Use when you need a read-only handoff summary of current branch state, local/remote refs, worktrees, active Flowser plans, selected-plan hints, and suggested next checks.
vc:audit-context
Audit Flowser context routing, shared-skill discoverability, and Codex/Codex wiring. Use when context docs or skill surfaces move, split, or drift.