baoyu-format-markdown

Pass

Audited by Gen Agent Trust Hub on Mar 17, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes a local TypeScript script (scripts/main.ts) using the Bun runtime or npx to apply typography improvements. It additionally spawns a subprocess to run the autocorrect-node utility for CJK spacing fixes.
  • [PROMPT_INJECTION]: The skill's workflow involves reading and analyzing untrusted user-provided content to generate titles, summaries, and structural improvements, which presents a surface for indirect prompt injection. 1. Ingestion points: User-specified files are read and processed in the Analysis phase (SKILL.md). 2. Boundary markers: The skill includes instructional constraints for the agent (e.g., 'Never add, delete, or rewrite content') to mitigate accidental changes, though it lacks cryptographic delimiters for user data. 3. Capability inventory: The skill can write files to the local filesystem and execute shell commands via its bundled scripts. 4. Sanitization: No explicit sanitization or filtering of the input content is observed before it is used to populate prompts for metadata generation.
  • [EXTERNAL_DOWNLOADS]: The skill utilizes npx to execute the autocorrect-node package, which may download the dependency from the npm registry at runtime if not already cached.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 17, 2026, 01:48 PM