security-audit
Pass
Audited by Gen Agent Trust Hub on Sep 30, 2026
Risk Level: SAFEDYNAMIC_CONTEXT_INJECTIONINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [DYNAMIC_CONTEXT_INJECTION]: The skill executes a shell command at load time to resolve project configuration. This command targets a local script within the project structure and is used for project-specific tooling as defined in the skill environment.
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze untrusted project source code and data files.
- Ingestion points: Processes all files in the resolved code root, including assets and configuration files.
- Boundary markers: Includes explicit instructions to never assume a pattern is safe and requiring human review for findings.
- Capability inventory: Spawns agents and executes shell-based search commands via Grep and Bash.
- Sanitization: Relies on search logic and agent analysis rather than data escaping.
- [COMMAND_EXECUTION]: The skill uses shell tools to perform code analysis. Tool usage is explicitly constrained to specific local scripts and parameters in the skill configuration.
Audit Metadata