avoid-ai-writing
Pass
Audited by Gen Agent Trust Hub on Aug 22, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill operates entirely on local prose patterns and explicitly forbids making inferences about authorship or external claims (SKILL.md).
- [SAFE]: The execution environment is restricted to local validation scripts (
scripts/check.py) that perform structural integrity checks on the package itself. These scripts use safe practices such as path resolution checks to prevent path traversal and do not have external dependencies. - [SAFE]: The skill does not perform any network operations, data exfiltration, or sensitive file access. The resources are limited to internal markdown reference files.
- [SAFE]: Instructions include clear boundary markers and safety guidelines, instructing the agent to preserve facts, citations, and terminology during edits, and to mark behavioral verifications as 'UNVERIFIED' (SKILL.md).
- [SAFE]: External URLs referenced in the documentation (
keepachangelog.com,arxiv.org,tropes.fyi) are well-known educational or industry-standard sources and are used neutrally for citation and context.
Audit Metadata