skills/fallomai/skills/crow/Gen Agent Trust Hub

crow

Pass

Audited by Gen Agent Trust Hub on Mar 14, 2026

Risk Level: SAFENO_CODECOMMAND_EXECUTIONPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [NO_CODE]: This skill is composed exclusively of Markdown documentation and instructional guides; it contains no executable code files or scripts.
  • [COMMAND_EXECUTION]: The instructions guide the agent to perform network operations using curl to interact with the service API at api.crowpay.ai for wallet setup and payment authorization.
  • [PROMPT_INJECTION]: The skill instructs agents to ingest and process the body of HTTP 402 "Payment Required" responses from external APIs. This creates a surface for indirect prompt injection as a malicious external service could embed instructions within the payment metadata.
  • Ingestion points: Data from external HTTP 402 responses is ingested into the agent context as described in SKILL.md and references/x402-flow.md.
  • Boundary markers: No explicit delimiters or instructions to ignore embedded commands are present in the documentation for handling the external JSON payload.
  • Capability inventory: The skill uses curl for making authorized network requests to the payment provider.
  • Sanitization: The documentation does not describe sanitization or validation of the external payment request object before forwarding it to the authorization endpoint.
  • [EXTERNAL_DOWNLOADS]: The documentation references the installation of additional tools from the vendor's GitHub repository (github.com/Fallomai/skills) using the npx package runner.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 14, 2026, 06:33 PM