macro-builder

Pass

Audited by Gen Agent Trust Hub on Jul 1, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructs the agent to execute composio CLI commands in Bash to search for and retrieve support tickets from the Gorgias platform. This behavior is aligned with the skill's stated purpose of support operations analysis.\n- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests and processes ticket subjects and messages from an external source (Gorgias).\n
  • Ingestion points: Support ticket data retrieved via composio execute GORGIAS_LIST_TICKETS in SKILL.md.\n
  • Boundary markers: None present; the skill does not use delimiters or instructions to ignore instructions embedded within the retrieved ticket content.\n
  • Capability inventory: The agent has access to bash and the composio toolset for data retrieval and execution.\n
  • Sanitization: There is no evidence of sanitization or filtering applied to the external ticket data before it is analyzed by the AI.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 1, 2026, 08:14 PM
Security Audit — agent-trust-hub — macro-builder