chromatic-turbosnap-check

Pass

Audited by Gen Agent Trust Hub on Oct 2, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill performs its core function by executing local commands including Git, the project's Storybook build script, and the Chromatic CLI. Commands are invoked via spawnSync with shell: false to prevent shell-based injection attacks. The instructions explicitly warn the agent to use only prepared or authorized configuration files to avoid executing arbitrary commands from untrusted repositories.\n- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted repository data including Git logs, source code, and build artifacts which could theoretically contain malicious instructions to influence the agent.\n
  • Ingestion points: Source files and stats files are read in scripts/preflight.mjs and scripts/preflight-lib.mjs.\n
  • Boundary markers: SKILL.md contains explicit instructions to treat repository data as data, not instructions.\n
  • Capability inventory: The skill uses spawnSync for repository interaction and writes reports to the local filesystem.\n
  • Sanitization: File paths are normalized and commands are executed without a shell. Source analysis is performed via a static parser.\n- [DYNAMIC_EXECUTION]: The skill includes a bundled version of @babel/parser to analyze runtime imports in JavaScript and TypeScript files. This analysis is static; the code is parsed into an AST for inspection of import/export statements rather than being executed or evaluated.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 2, 2026, 08:30 PM