creating-kb

Pass

Audited by Gen Agent Trust Hub on Jun 25, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The test_parity.py script uses subprocess.run to execute local node and python3 searchers to ensure consistent behavior across runtimes. These calls are confined to testing with controlled inputs and local file paths.
  • [PROMPT_INJECTION]: The skill facilitates retrieval-augmented generation (RAG) by packaging files into a searchable index. The bundle template (bundle_SKILL.md) instructs the consuming agent to prioritize retrieved chunks as authoritative context. While the ingestion process does not filter for potential instructions inside user-provided documents, this represents a standard retrieval surface rather than a malicious injection within the skill's own logic.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 25, 2026, 04:58 PM
Security Audit — agent-trust-hub — creating-kb