fabric-cli-core
Pass
Audited by Gen Agent Trust Hub on Oct 18, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- Command Execution for CLI Interaction: The skill utilizes shell commands through the
fabCLI and a Python script (scripts/health_check.py) to perform its primary functions. The Python script employs safe subprocess execution patterns to verify the CLI environment, authentication status, and resource connectivity. - Potential for Indirect Prompt Injection: By design, the skill reads and processes external data from Fabric resources, such as notebook definitions and semantic model metadata. While this creates a surface for indirect prompt injection if those resources contain malicious instructions, the skill incorporates explicit rules to validate JSON inputs and sanitize file paths to mitigate these considerations.
- Sensitive Data Handling Practices: The skill provides documentation on the location of CLI configuration and credential caches (e.g.,
~/.config/fab/auth.json). It maintains a secure posture by instructing the agent to never output or request sensitive strings like tokens or secrets, and instead promotes secure authentication patterns.
Audit Metadata