proposal-draft

Pass

Audited by Gen Agent Trust Hub on Jul 1, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes the Bash shell to execute commands for tool discovery and proposal creation through the composio CLI.
  • [PROMPT_INJECTION]: The skill demonstrates an indirect prompt injection surface (Category 8) by ingesting untrusted data and utilizing powerful shell capabilities.
  • Ingestion points: The skill accepts input via the $ARGUMENTS variable and instructions in SKILL.md suggest pulling context from HubSpot deals or support tickets.
  • Boundary markers: There are no delimiters or specific instructions for the agent to distinguish between its own logic and potentially malicious content within the external data.
  • Capability inventory: The skill possesses shell execution capabilities via Bash, which are used to search for tools and execute proposal creation actions.
  • Sanitization: There is no evidence of sanitization or escaping of the ingested data before it is interpolated into the proposal draft or the JSON payload for the shell command.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 1, 2026, 08:13 PM
Security Audit — agent-trust-hub — proposal-draft