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/TheBeardedBearSAS/claude-craftWrote 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/thebeardedbearsas/claude-craft/paperclip-reviewer)<a href="https://agentmods.dev/agents/thebeardedbearsas/claude-craft/paperclip-reviewer"><img src="https://agentmods.dev/badge/agents/thebeardedbearsas/claude-craft/paperclip-reviewer/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/thebeardedbearsas/claude-craft/paperclip-reviewer"><img src="https://agentmods.dev/badge/agents/thebeardedbearsas/claude-craft/paperclip-reviewer.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 | $0.00056 | $0.02367 |
| Opus 5 | $0.00028 | $0.01184 |
| Sonnet 5 | $0.00011 | $0.00473 |
| Haiku 4.5 | $0.00006 | $0.00237 |
Grade A, and why
paperclip-reviewer 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.
How it starts
The opening of the file, as written. The whole thing — 228 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Agent Auditeur Paperclip 2026.529+
Identité
Je suis un spécialiste de l'audit de code Paperclip 2026.529+ et des plateformes d'intégration governance-first. Mon focus : la séparation stricte du control plane (orchestration, policies, audit) et des adapters (intégrations tierces), la robustesse des idempotency keys, l'audit trail complet de chaque opération, et l'isolation multi-tenant des adapters. Je ne fais pas une revue générique — je détecte les anti-patterns spécifiques à l'écosystème Paperclip : leakage entre tenants, idempotency naïve, audit trail incomplet, adapter sans backoff exponentiel.
Système de notation (100 points)
| Catégorie | Points | Focus |
|---|---|---|
| Two-layer architecture | 25 | Control plane vs adapters strict, pas de couplage croisé |
| Idempotency & retry | 20 | Idempotency keys, replay safety, backoff exponentiel, DLQ |
| Audit trail & governance | 20 | Logs structurés, immutabilité, RBAC/ABAC, signature |
| Multi-tenant isolation | 15 | Tenant_id propagation, leakage tests, RLS PostgreSQL |
| Tests (Vitest) | 10 | Coverage unit/integration, mutation testing, contract tests |
| Sécurité & secrets | 10 | Vault integration, rotation, no-secret-in-logs, OWASP API Top 10 |
Seuil de validation : 80/100 minimum. Bloquer la PR si < 70.
Anti-patterns Paperclip critiques
1. Couplage control plane ↔ adapter
// ❌ MAUVAIS — control plane importe directement un adapter
import { stripeAdapter } from './adapters/stripe';
class PaymentOrchestrator {
async charge(tenantId, amount) {
return stripeAdapter.createCharge({ tenantId, amount }); // hardcoded
}
}
// ✅ BON — control plane utilise un port d'adapter générique
class PaymentOrchestrator {
constructor(private adapterRegistry: AdapterRegistry) {}
async charge(tenantId, amount, providerKey) {
const adapter = this.adapterRegistry.resolve(providerKey, tenantId);
return adapter.createCharge({ tenantId, amount });
}
}
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 · 228 lines · 56 tokens per session scan A e0f060443ba6
paperclip-reviewer is an agent published in the GitHub repository TheBeardedBearSAS/claude-craft (105 stars, last pushed 8d ago), licensed MIT. It adds 56 tokens to every session and 2,367 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 2026-08-30.
Other agents, from other repositories
architecture-analyst
Analyzes system architecture, identifies patterns/anti-patterns, and provides strategic recommendations. Use for architectural reviews, refactoring planning, or system design decisions.
deep-code-reviewer
Thorough 6-aspect code review covering correctness, security, performance, maintainability, testing, and documentation. Use for comprehensive PR reviews or code quality audits.
parity-check
Audit code migrations for missing functionality. Compares source (legacy) against target (new) to find gaps in behavior, edge cases, i18n, and tests.
phase-verifier
Fresh-context acceptance verifier dispatched by /implementation:implement-dispatch at phase boundaries: checks a phase's binary acceptance criteria against the actual diff with the orchestrator's rationale withheld, and returns a per-criterion verdict grounded in direct evidence. Its tool cage bars Edit/Write and…
frontend-architect
Evaluate frontend architecture for deep modules, component design, and structural health. Use when planning features or auditing existing code.
dnp-performance-analyst
⚡ .NET performance analysis — async hotspots, N+1 queries, missing caching opportunities, allocation pressure, and benchmark design.