skills/richardbray/skills/title-score/Gen Agent Trust Hub

title-score

Pass

Audited by Gen Agent Trust Hub on Apr 29, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill performs local text processing using standard Python libraries (argparse, json, re, sys). There are no network requests or external data transfers detected.
  • [COMMAND_EXECUTION]: While the skill uses a Python script, it is a self-contained utility for string manipulation and scoring. It does not spawn dangerous subprocesses or execute arbitrary shell commands.
  • [DATA_EXFILTRATION]: No evidence of data exfiltration. The script processes input titles and returns scores without accessing sensitive environment variables or local files like SSH keys or credentials.
  • [PROMPT_INJECTION]: The instructions are focused on title scoring and do not contain patterns designed to bypass agent safety filters or override system instructions.
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests user-provided titles for scoring. However, the processing is limited to regex-based word extraction and length calculations, which do not provide an execution path for malicious instructions contained within the data.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 29, 2026, 07:55 PM