dunsocial

Pass

Audited by Gen Agent Trust Hub on Jul 30, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill documents the use of the dun CLI tool to perform social media management tasks.\n- [EXTERNAL_DOWNLOADS]: The skill requires downloading the dunsocial NPM package, which is the official package for the service.\n- [CREDENTIALS_UNSAFE]: The skill uses environment variables for authentication with safe documentation placeholders (dun_pat_...).\n- [INDIRECT_PROMPT_INJECTION]: The skill facilitates the ingestion of external data via workspace memory and file uploads.\n
  • Ingestion points: dun memory search and dun posts schedule-thread --file ./thread.json in SKILL.md.\n
  • Boundary markers: None present.\n
  • Capability inventory: Network API access, local file reading, and CLI execution.\n
  • Sanitization: Not specified.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 30, 2026, 06:39 PM
Security Audit — agent-trust-hub — dunsocial