llm-wiki-management

Pass

Audited by Gen Agent Trust Hub on Jul 11, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The Python scripts lint_wiki.py, check_wiki_fixtures.py, and ingest_diff.py utilize subprocess.run to execute local git commands and invoke other scripts within the package. These calls are used for verifying repository status, commit metadata, and running automated fixture checks as part of the wiki maintenance workflow. These operations do not use shell=True and target the local workspace only.
  • [SAFE]: The skill processes potentially untrusted user data from the raw/ directory to generate summaries and synthesis pages. While this represents a surface for indirect prompt injection, the skill mitigates this through a disciplined workflow that requires human review markers (reviewed: true) for verified content and enforces a strict boundary between user-controlled raw data and agent-controlled wiki assets.
  • [SAFE]: No malicious obfuscation, data exfiltration, or remote code execution patterns were identified. The skill relies on standard library Python modules and avoids external dependencies to ensure a secure, local-first execution environment.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 11, 2026, 09:51 AM
Security Audit — agent-trust-hub — llm-wiki-management