ast-grep

Pass

Audited by Gen Agent Trust Hub on Sep 19, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill's installation scripts (install.sh and install.ps1) are designed to download prebuilt binaries of the ast-grep tool from its official GitHub releases page. This is a standard and expected mechanism for establishing the skill's core functionality.
  • [COMMAND_EXECUTION]: The Python wrapper (scripts/ast_grep_helper.py) executes the ast-grep CLI tool using the subprocess.run method. It passes arguments as a list rather than a single string, which effectively prevents shell injection vulnerabilities. The script also includes a 'doctor' command to verify the binary's integrity and version.
  • [INDIRECT_PROMPT_INJECTION]: As a code search and modification tool, the skill processes untrusted files within a user's project. While it presents matched code snippets to the agent, the risk of indirect prompt injection is inherent to the tool's purpose of reading external code. The skill's use of structural matching rather than raw text matching reduces the likelihood of accidentally executing instructions embedded in comments or strings, and the surface area is typical for this category of developer tool.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 19, 2026, 06:00 PM