clari-common-errors

Pass

Audited by Gen Agent Trust Hub on Sep 13, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests data from external Clari API endpoints, which represents a standard potential attack surface for indirect prompt injection if the external service were compromised.
  • Ingestion points: Data retrieved from api.clari.com and api-doc.copilot.clari.com (SKILL.md).
  • Boundary markers: The instructions do not specify explicit delimiters or markers to isolate the API data from the agent's reasoning context.
  • Capability inventory: The skill is limited to Read, Grep, and Bash(curl:*) as specified in the allowed-tools configuration.
  • Sanitization: While the skill instructs the agent to redact incident records and keep tokens out of chat, it does not explicitly define sanitization for the incoming API payload before it is processed by the agent.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 13, 2026, 01:17 PM
Security Audit — agent-trust-hub — clari-common-errors