cursor-correct
Pass
Audited by Gen Agent Trust Hub on Oct 20, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to analyze historical repository data, which could contain instructions intended to influence the agent's behavior.
- Ingestion points: Recent commits, reverts, review comments, and agent instruction files (as specified in SKILL.md).
- Boundary markers: The instructions lack explicit delimiters or warnings to ignore instructions embedded within the historical data being analyzed.
- Capability inventory: The agent is authorized to modify repository files, including architecture, types, and tests, and to execute shell commands to verify these fixes (SKILL.md).
- Sanitization: No explicit sanitization or validation is performed on the historical data before it is used to determine repository modifications.
- [COMMAND_EXECUTION]: The skill requires the agent to execute commands locally to verify that mistake-prevention mechanisms work as intended.
- Evidence: The instructions in SKILL.md explicitly state: "Run the same command locally and in CI" to prove checks fail on real past mistakes.
Audit Metadata