astrofy-checkup
Pass
Audited by Gen Agent Trust Hub on Oct 12, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructions specify the use of the 'astrofy' CLI tool for status checks, audit reports, and execution of validation rules.
- [INDIRECT_PROMPT_INJECTION]: The skill reads from local configuration files which serves as an ingestion point for potentially untrusted data.
- Ingestion points: Configuration files such as policies.json, checks.json, and paths.json are ingested during the execution of status and check commands.
- Boundary markers: No delimiters or 'ignore embedded instructions' warnings are provided to the agent.
- Capability inventory: The skill possesses shell execution (astrofy CLI), file system read, and file system write capabilities.
- Sanitization: No explicit sanitization or escaping of file content is defined before the data enters the agent context.
- [DYNAMIC_EXECUTION]: The skill facilitates the execution of local scripts through a 'trustedExecution' configuration flag, which permits the tool to run project-specific scripts at runtime based on external configuration.
Audit Metadata