clawhub-content-rights-correspondence

Pass

Audited by Gen Agent Trust Hub on Jul 10, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes administrative commands using bun run admin for email transmission and record management within the local environment as part of its core functionality.
  • [PROMPT_INJECTION]: The skill features an indirect prompt injection surface due to the inclusion of user-controlled variables in shell command arguments.
  • Ingestion points: User-supplied publisher handles and email content as specified in the correspondence workflows.
  • Boundary markers: Safety rules explicitly require the agent to perform a dry-run and obtain user sign-off on the final draft before sending.
  • Capability inventory: Shell command execution capabilities are present via the bun run admin tool mentioned in SKILL.md.
  • Sanitization: The instructions do not define specific sanitization or validation routines for external inputs before CLI usage.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 10, 2026, 03:30 PM
Security Audit — agent-trust-hub — clawhub-content-rights-correspondence