Sécurité Paperclip — isolation tenant, secrets, portes d'approbation, budgets stricts, capacités plugin minimales. À utiliser pour auditer ou durcir Paperclip.
Scanned 9/5/2026
Install to Claude Code
Skills Directorynpx -y skills add TheBeardedBearSAS/claude-craft --skill security-paperclip --agent claude-codeInstalls into .claude/skills of the current project.
Are you the author of Security Paperclip?
Add the live security badge to your README — it updates automatically with every re-scan.
[](https://www.skillsdirectory.com/skills/thebeardedbearsas-security-paperclip-177bbbac)More formats (shields.io, HTML) on the badges page.
---
name: security-paperclip
description: Sécurité Paperclip — isolation tenant, secrets, portes d'approbation, budgets stricts, capacités plugin minimales. À utiliser pour auditer ou durcir Paperclip.
---
# Sécurité Paperclip
`companyId` tiré uniquement de la session/du path (jamais du body client) ; secrets chiffrés au repos + rédigés dans les logs + résolus via `ctx.secrets.resolve(ref)` dans les plugins ; portes d'approbation serveur-only et append-only ; les budgets sont des limites strictes appliquées au dispatch ; Better Auth pour l'auth opérateur avec un `BETTER_AUTH_SECRET` tourné ; CSP/HSTS/COOP/CORP livrés sur l'UI ; capacités de plugin déclarées au minimum ; `pnpm audit --audit-level=high` dans la CI.
Voir ../../rules/11-security-paperclip.md pour la documentation détaillée.
Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.
No comments yet. Be the first to comment!
Adversarial multi-agent planning skill. Self-orchestrates 5 hostile category members (unspecified-low, unspecified-high, deep, ultrabrain, artistry) via team-mode for ruthless cross-critique debate, distills only the defensible insights, then MANDATORILY hands the distilled insight bundle to the `plan` agent for executable plan formalization. Use when planning needs maximum rigor and surfacing of weak assumptions, blind spots, and over-engineering. Triggers: 'hyperplan', 'hpp', '/hyperplan', ...
