datahub-mfe-configure-app

Pass

Audited by Gen Agent Trust Hub on Apr 1, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes local scripts scripts/dev/datahub-dev.sh and run-local-frontend (located in datahub-frontend/run/) to rebuild and restart the DataHub frontend after configuration updates.
  • [PROMPT_INJECTION]: The skill has an indirect prompt injection surface where untrusted data is processed. 1. Ingestion points: User-provided configuration values enter via the AskQuestion tool in SKILL.md. 2. Boundary markers: Delimiters or 'ignore embedded instructions' warnings are absent when interpolating input into configuration files. 3. Capability inventory: The skill performs file-write operations on configuration YAMLs and executes shell scripts across SKILL.md. 4. Sanitization: No escaping or validation of external content is performed before interpolation.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 1, 2026, 03:25 PM