json-config-manager

Pass

Audited by Gen Agent Trust Hub on Feb 19, 2026

Risk Level: SAFENO_CODE
Full Analysis
  • NO_CODE (SAFE): The provided skill contains only Markdown documentation (SKILL.md) and no executable scripts, binaries, or configuration files. No security threats are present in the provided content.
  • INDIRECT_PROMPT_INJECTION (SAFE): The skill is intended to manage JSON configurations, implying ingestion of external data. While the use of powerful tools like 'Bash' and 'Read' creates a potential attack surface, the absence of implementation logic makes this risk negligible in its current state. Evidence: 1. Ingestion points: JSON configuration files. 2. Boundary markers: None. 3. Capability inventory: Bash, Read, Write, Edit, Grep. 4. Sanitization: None.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 19, 2026, 12:46 AM