Skills
skills/roger-235/my_skill_repo/content-production/Gen Agent Trust Hub

content-production

Pass

Audited by Gen Agent Trust Hub on Apr 5, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTIONNO_CODE
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructs the agent to run a local script named scripts/content_scorer.py to evaluate draft readability.\n- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it processes external data from marketing-context.md and user-provided topics without applying sanitization or boundary markers.\n
  • Ingestion points: User input (topics, keywords) and the marketing-context.md file.\n
  • Boundary markers: None present in the prompt templates or instructions.\n
  • Capability inventory: File system access (reading context) and local command execution (scripts/content_scorer.py).\n
  • Sanitization: No input validation or content filtering is specified.\n- [NO_CODE]: The Python script scripts/content_scorer.py, which is central to the optimization phase, is not included in the provided file set, rendering its logic unverifiable.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 5, 2026, 08:05 PM
Security Audit — agent-trust-hub — content-production