verify
Pass
Audited by Gen Agent Trust Hub on Oct 22, 2026
Risk Level: SAFECOMMAND_EXECUTIONDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill explicitly instructs the agent to execute shell commands, test runners, and networking tools like curl to verify application behavior and record evidence.
- [DYNAMIC_EXECUTION]: To gather evidence for behaviors without existing entry points, the skill directs the agent to generate and run scratch scripts or temporary tests outside the production codebase.
- [INDIRECT_PROMPT_INJECTION]: The skill processes project-specific data (AGENTS.md, specifications, diffs) and external network responses to determine verification claims, which serves as an ingestion point for untrusted content.
- Ingestion points: repository configuration files, user-provided specifications, commit history, and external API/web content retrieved via curl or browser automation.
- Boundary markers: The instructions do not define specific delimiters or isolation protocols for data retrieved from the environment.
- Capability inventory: High-level access including full shell command execution, file system read/write, and network operations.
- Sanitization: No automated sanitization of ingested content is described; the skill relies on the agent's interpretation of the project context.
Audit Metadata