query
Pass
Audited by Gen Agent Trust Hub on Oct 1, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill fetches and processes data from the Stacks blockchain via the Hiro API, which represents a surface for indirect prompt injection if transaction memos or contract events contain malicious instructions.
- Ingestion points: Data retrieved from Hiro API endpoints (blocks, transactions, events) in
query.ts. - Boundary markers: The skill does not explicitly use delimiters or warnings to isolate blockchain-sourced data from the agent's instructions.
- Capability inventory: The skill is limited to read-only network queries and CLI output; it lacks file-write or destructive network capabilities.
- Sanitization: CLI arguments for contract calls are validated using
JSON.parseand Clarity-specific hex conversion. - [EXTERNAL_DOWNLOADS]: Connects to the Hiro API (
api.hiro.so) to retrieve blockchain information. This is a well-known service for Stacks network developers and users.
Audit Metadata