skills/openlark/skills/gray-matter/Gen Agent Trust Hub

gray-matter

Warn

Audited by Gen Agent Trust Hub on Aug 8, 2026

Risk Level: MEDIUMEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The file scripts/parse.js contains logic to automatically fetch and install the gray-matter library from the NPM registry if it is not found locally.
  • [COMMAND_EXECUTION]: The skill performs shell command execution using child_process.execSync in scripts/parse.js to manage package dependencies at runtime. It also reads from user-provided file paths via fs.readFileSync.
  • [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface (Category 8). Ingestion points: Data is read from external files or stdin in scripts/parse.js. Boundary markers: Uses delimiters like --- or ~~~ to identify content. Capability inventory: Uses execSync and require for dynamic operations. Sanitization: No input validation or sanitization is performed on the parsed content before it is returned as JSON or YAML output. Maliciously crafted front-matter in processed documents could contain instructions that influence the agent's behavior when the parsed output is returned to the context.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Aug 8, 2026, 04:17 AM
Security Audit — agent-trust-hub — gray-matter