read-x-links-vercel

Pass

Audited by Gen Agent Trust Hub on Jun 29, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill communicates with an external API at x.pcstyle.dev to perform its primary function of converting X posts to Markdown.
  • Evidence: scripts/read-x.sh executes curl requests to https://x.pcstyle.dev/api/convert.
  • Context: The target domain is a vendor resource belonging to the author pc-style.
  • [PROMPT_INJECTION]: The skill is subject to indirect prompt injection risks as it incorporates untrusted content from social media posts into the agent's context.
  • Ingestion points: The output of scripts/read-x.sh which contains the text of X posts.
  • Boundary markers: Absent; the fetched content is not wrapped in specific delimiters to distinguish it from instructions.
  • Capability inventory: The skill employs Bash and curl for its operations.
  • Sanitization: Absent; the script returns the raw Markdown or JSON body received from the API.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 29, 2026, 10:32 AM
Security Audit — agent-trust-hub — read-x-links-vercel