ast-grep
Pass
Audited by Gen Agent Trust Hub on Sep 6, 2026
Risk Level: SAFEDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The content is entirely focused on providing documentation and workflows for the ast-grep structural search tool. No malicious code, obfuscation, or unauthorized access patterns were found.
- [DYNAMIC_EXECUTION]: The skill's workflow includes generating temporary test files and executing the
ast-grepCLI tool to validate patterns, which is standard for code analysis tasks. - [INDIRECT_PROMPT_INJECTION]: The skill ingests user input to generate ast-grep patterns. It addresses potential injection risks by instructing the agent to escape metavariables (
\$VAR) when using the CLI. Evidence: 1. Ingestion points: User queries for code patterns in SKILL.md; 2. Boundary markers: YAML rule structure and CLI argument separation; 3. Capability inventory:ast-grepCLI execution for file reading; 4. Sanitization: Mandatory shell variable escaping instructions.
Audit Metadata