buy-crypto
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes repository context to assist with integration, establishing a potential surface for indirect prompt injection where instructions could be embedded in codebase files.\n
- Ingestion points: User repository files and configuration (SKILL.md).\n
- Boundary markers: Strict instruction hierarchy prioritizes user commands over repository context (SKILL.md).\n
- Capability inventory: Network calls to Circle API endpoints for session minting (references/app-kit.md, references/onramp-kit.md).\n
- Sanitization: Security rules mandate authentication, server-side address validation, and on-chain verification (SKILL.md).\n- [COMMAND_EXECUTION]: Documentation includes standard shell commands for environment setup and package installation.\n
- Evidence: Node.js version management (
nvm use 22) and SDK installation (npm install).\n- [EXTERNAL_DOWNLOADS]: Fetches vendor SDKs and references infrastructure for API and widget services.\n - Evidence: Downloads
@circle-fin/app-kitand@circle-fin/onramp-kitfrom the public npm registry; referencesapi.circle.comandonramp.arc.io.
Audit Metadata