exa-enterprise-rbac

Pass

Audited by Gen Agent Trust Hub on Mar 24, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: The skill facilitates the ingestion of external data from the Exa search API, creating a surface for indirect prompt injection if web content returned by the API contains malicious instructions.
  • Ingestion points: Search results from api.exa.ai/search (SKILL.md).
  • Boundary markers: Absent in implementation examples.
  • Capability inventory: Write and Edit tools are permitted in the skill configuration.
  • Sanitization: No explicit validation or filtering of API response content is provided in the examples.
  • [COMMAND_EXECUTION]: Provides examples using curl and jq to interact with the Exa API. These commands follow security best practices by utilizing environment variables ($EXA_ADMIN_KEY, $EXA_API_KEY) for authentication rather than hardcoded secrets. The use of the official api.exa.ai domain is consistent with the skill's stated purpose for enterprise RBAC management.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 24, 2026, 04:52 PM