bug-reporting
Pass
Audited by Gen Agent Trust Hub on Sep 16, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process data from external, potentially untrusted sources such as automated accessibility tool outputs and user reports. This represents a vulnerability surface for indirect prompt injection.
- Ingestion points: The skill is intended to process 'automated tool output, manual testing, user reports, or testing with disabled people' as defined in SKILL.md.
- Boundary markers: Instructions require the agent to 'Record observed facts separately from assumptions' and to label the 'evidence basis' clearly, which provides context separation.
- Capability inventory: The skill does not provide the agent with dangerous capabilities such as arbitrary command execution, network exfiltration, or file system access.
- Sanitization: The skill contains multiple mandatory instructions to 'remove personal data, credentials, tokens, and private content' from all reports and attachments, significantly mitigating the risks associated with data processing.
Audit Metadata