write-talk-script

Pass

Audited by Gen Agent Trust Hub on Jul 7, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill utilizes local Python scripts (word_budget.py and script_timer.py) that are explicitly restricted to the Python standard library and contain no network or file-system-modifying capabilities beyond reading specified input files.
  • [SAFE]: Explicit guardrails are present to prevent the AI from fabricating data (hallucination) or performing unauthorized actions, such as submitting documents to external portals on the user's behalf.
  • [SAFE]: No patterns of prompt injection, obfuscation, or unauthorized data access were found in the instructional content or the supporting scripts.
  • [SAFE]: Indirect prompt injection risks are mitigated by the specific task focus (word counting and spoken prose generation) and instructions to use the user's paper as the sole source of truth.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 7, 2026, 01:02 PM
Security Audit — agent-trust-hub — write-talk-script