datahub-connector-planning

Pass

Audited by Gen Agent Trust Hub on Mar 7, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [SAFE]: No malicious patterns were identified in the skill. The logic follows a structured workflow for connector design.
  • [PROMPT_INJECTION]: The skill has an attack surface for indirect prompt injection because it reads untrusted data from the web during research. 1. Ingestion points: WebSearch/WebFetch tools in SKILL.md. 2. Boundary markers: Absent. 3. Capability inventory: Bash (ls, find, grep, pip index) and sub-agent dispatch. 4. Sanitization: Absent.
  • [COMMAND_EXECUTION]: The skill uses restricted Bash commands for discovery tasks like identifying similar connectors and checking library availability, which is legitimate for its purpose.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 7, 2026, 02:55 PM