acp-capability-negotiation
Pass
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill fetches protocol documentation and JSON schemas from GitHub and OpenAI's developer portal. These are recognized, well-known technology services, and the documentation is intended to support the implementation of the Agentic Commerce Protocol.
- [INDIRECT_PROMPT_INJECTION]: The skill has a potential vulnerability surface as it ingests untrusted data from external websites via web tools.
- Ingestion points: SKILL.md (instructions to search GitHub and fetch content from openai.com).
- Boundary markers: The skill lacks explicit boundary markers or instructions to the agent to disregard instructions embedded in the fetched content.
- Capability inventory: The skill is configured with access to powerful tools like Bash, Write, and Edit which could be misused if malicious content is processed.
- Sanitization: There is no evidence of validation or sanitization mechanisms for the data retrieved from the web.
Audit Metadata