zero-slop
Pass
Audited by Gen Agent Trust Hub on Nov 10, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes a bundled Python script (
scripts/slopscore.py) to perform prose scoring and factual fidelity checks. - [INDIRECT_PROMPT_INJECTION]:
- Ingestion points: The skill processes user-supplied text drafts (untrusted data).
- Boundary markers:
SKILL.mdexplicitly instructs the agent to treat drafts as data and ignore embedded instructions. - Capability inventory: The skill executes shell commands via a local Python script and reads/writes local files.
- Sanitization: No active sanitization of the input text, but the script performs non-executable regex-based analysis.
- [SAFE]: The skill operates locally, avoids external network calls, and includes privacy-focused design that prevents persistent state unless explicitly requested by the user.
Audit Metadata