acp-discount-extension
Pass
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill's primary purpose is to provide implementation guidance for a standardized commerce protocol. The instructions are technical and transparent.
- [EXTERNAL_DOWNLOADS]: The skill directs the agent to fetch documentation from trusted and well-known sources, including official developer documentation at developers.openai.com and specific protocol search queries on GitHub.
- [INDIRECT_PROMPT_INJECTION]: The skill defines a data ingestion surface by instructing the agent to fetch external specifications.
- Ingestion points: External technical documentation from GitHub and OpenAI.
- Boundary markers: None explicitly defined for the fetched content.
- Capability inventory: The skill allows use of tools such as Bash, Write, and WebFetch, which are necessary for implementing the protocol based on the fetched specifications.
- Sanitization: No specific sanitization logic is provided for the fetched documentation strings.
- Note: This is an inherent property of documentation-fetching skills and represents no active threat in this context.
Audit Metadata