cursor-bro
Pass
Audited by Gen Agent Trust Hub on Oct 19, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructions require the agent to process the conversation history ("restating the last message") without employing specific boundary markers or delimiters to isolate the untrusted data from the agent's instructions.
- Ingestion points: Agent conversation history accessed via prompt instructions (SKILL.md).
- Boundary markers: Not present.
- Capability inventory: The skill does not utilize high-risk tools or capabilities such as network requests, file writing, or command execution.
- Sanitization: No evidence of input filtering or escaping for the ingested message content.
- [EXTERNAL_DOWNLOADS]: The skill documentation (MIRROR.md) references an external GitHub repository belonging to Cursor (github.com/cursor/plugins.git) for the purposes of source tracking, version pinning, and providing a physical source path for the mirrored content.
Audit Metadata