better-interface
Pass
Audited by Gen Agent Trust Hub on Oct 24, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to read and process external project documentation such as
CONTRIBUTING.md,CODING_STANDARDS.md,AGENTS.md, andCLAUDE.md, alongside source code files. These files represent untrusted data that could contain instructions intended to override the agent's behavior. - Ingestion points: Processes
CONTRIBUTING.md,CODING_STANDARDS.md,AGENTS.md,CLAUDE.md, and project source files (implied by location citations). - Boundary markers: Absent; there are no specific delimiters or instructions to the agent to ignore instructions embedded within the ingested data.
- Capability inventory: The skill can execute project-specific commands (verification checks) and has the capability to modify source code if requested by the user.
- Sanitization: Absent; the skill does not specify any validation or filtering of the content read from the files.
- [COMMAND_EXECUTION]: The skill instructs the agent to 'Run the safe, relevant checks the project offers' and inspect the rendered interface. If the project being reviewed contains malicious test scripts or build commands, the agent could be induced to execute harmful code during the verification phase.
Audit Metadata