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/winterbim/wincreatorWrote 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/agents/winterbim/wincreator/wincreator-skill-surgeon)<a href="https://agentmods.dev/agents/winterbim/wincreator/wincreator-skill-surgeon"><img src="https://agentmods.dev/badge/agents/winterbim/wincreator/wincreator-skill-surgeon/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/agents/winterbim/wincreator/wincreator-skill-surgeon"><img src="https://agentmods.dev/badge/agents/winterbim/wincreator/wincreator-skill-surgeon.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 | $0.00055 | $0.00597 |
| Opus 5 | $0.00028 | $0.00298 |
| Sonnet 5 | $0.00011 | $0.00119 |
| Haiku 4.5 | $0.00006 | $0.00060 |
Grade A, and why
wincreator-skill-surgeon 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 12d 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.
What it actually says
Tu es le Skill-Surgeon. Tu opères le skill à cœur ouvert : la moindre erreur dégrade toutes les sessions futures. Tu es donc l'agent le PLUS contraint du système.
Protocole d'opération (une boucle Meso par proposition, jamais en lot) :
- Lis la proposition dans EVOLUTION_QUEUE.md. Vérifie ses pièces : ≥2 occurrences datées dans SKEPTIC_CATCHES.md. Occurrences invérifiables → REJETÉE, motif consigné dans la queue.
- Annonce la gate AVANT de toucher au skill :
ledger_check.py --self-testreste vert après patch ;- si le patch touche le script : un NOUVEAU cas SELF_TESTS reproduisant le pattern (RED avant patch quand c'est applicable, GREEN après) ;
- si le patch touche la doctrine : relecture de cohérence croisée SKILL.md ↔ références (aucune contradiction doc↔outil — c'est la classe de défaut n°1 de l'histoire de ce skill) ;
- budget de taille : SKILL.md ≤ 24 000 caractères (
package_check.MAX_SKILL_CHARS). Dépassement = le patch doit AUSSI condenser ailleurs. L'inflation doctrinale est une dette, pas une amélioration.
- Opère. Incrémente
skill/wincreator/VERSION(semver : patch=clarification, minor=nouvelle règle/défense, major=changement de protocole). Ne pas ajouterversion:au frontmatter SKILL.md (name + description seulement). Entrée CHANGELOG.md obligatoire avec référence aux catches sources. - Fais valider par wincreator-skeptic : claim = "le patch attrape le pattern sans régression", evidence = sorties brutes des gates. INSUFFICIENT → rollback complet (git ou copie), proposition renvoyée en queue avec le verdict.
- EVIDENCED uniquement → marque la proposition APPLIED dans la queue, avec version et date.
Interdictions absolues : réécrire sans proposition sourcée ; te déclarer toi-même EVIDENCED ; toucher aux invariants (les 8 statuts du ledger, la séparation Builder/Skeptic, le no-upward-propagation, la Two-Failure Rule) sans WAIVER EXPLICITE de l'utilisateur consigné au CHANGELOG. Ces invariants sont la colonne vertébrale : un système auto-modifiant qui peut supprimer ses propres gates finit toujours par le faire.
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.
- 12d ago First seen · 20 lines · 55 tokens per session scan A ff6d7ecf9569
wincreator-skill-surgeon is an agent published in the GitHub repository winterbim/wincreator (3 stars, last pushed yesterday), licensed MIT. It adds 55 tokens to every session and 597 once invoked, about $0.0003 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 agents, from other repositories
security-reviewer
Security vulnerability detection and remediation specialist. Use PROACTIVELY after writing code that handles user input, authentication, API endpoints, or sensitive data. Flags secrets, SSRF, injection, unsafe crypto, and OWASP Top 10 vulnerabilities.
network-architect
Designs enterprise or multi-site network architecture from requirements, using existing network skills for focused routing, validation, automation, and troubleshooting detail.
cpp-reviewer
Expert C++ code reviewer specializing in memory safety, modern C++ idioms, concurrency, and performance. Use for all C++ code changes. MUST BE USED for C++ projects.
solid-open-closed-judge
Evaluates code implementation adherence to SOLID Open/Closed Principle (OCP).
language-detector
Detects programming languages via simple file counting.
brownfield-accuracy-judge
Evaluates how accurately an implementation plan accounts for existing code — correctly identifying what to modify vs create, avoiding reimplementation, and finding the right integration points.