check-agent-compatibility

Pass

Audited by Gen Agent Trust Hub on Aug 17, 2026

Risk Level: SAFECOMMAND_EXECUTIONREMOTE_CODE_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The workflow incorporates a compatibility scan step that executes a CLI tool via compatibility-scan-review to analyze repository metrics.
  • [REMOTE_CODE_EXECUTION]: The skill instructs the agent to boot the repository and verify changes during the startup-review and validation-review phases, which involves execution of untrusted code found within the repository.
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it reads and executes untrusted repository data. * Ingestion points: The repository files provided for the compatibility pass as described in SKILL.md. * Boundary markers: The instructions lack explicit delimiters or isolation commands to prevent the agent from obeying instructions embedded within audited files. * Capability inventory: In SKILL.md, the agent has access to tools for CLI execution and arbitrary code execution through the booting and validation steps. * Sanitization: No sanitization or filtering of the repository content is specified prior to execution.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 17, 2026, 11:52 AM
Security Audit — agent-trust-hub — check-agent-compatibility