acp-orders-webhooks

Pass

Audited by Gen Agent Trust Hub on Mar 31, 2026

Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [PROMPT_INJECTION]: The skill utilizes external data ingestion, which establishes an indirect prompt injection surface where malicious instructions in fetched documentation could potentially influence agent behavior.
  • Ingestion points: Documentation and implementation examples are retrieved via WebSearch and WebFetch from external sources like GitHub, OpenAI, and Stripe (SKILL.md).
  • Boundary markers: No explicit delimiters or instructions to ignore embedded commands within the fetched content are provided.
  • Capability inventory: The skill's environment allows high-privilege tools including Bash, Write, and Edit (SKILL.md).
  • Sanitization: The skill lacks instructions for sanitizing or validating the content retrieved from external sources.
  • [EXTERNAL_DOWNLOADS]: The skill directs the agent to fetch resources from external platforms, specifically OpenAI, Stripe, and the 'agentic-commerce-protocol' GitHub organization, for the purpose of guiding the implementation of the commerce protocol.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 31, 2026, 06:09 AM