brain-link-discipline
Pass
Audited by Gen Agent Trust Hub on Oct 13, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill directs the agent to use several shell commands to derive repository paths and verify remote availability, specifically using git, curl, and the gbrain CLI utility.
- [INDIRECT_PROMPT_INJECTION]: The skill processes and relays results from sub-agents, which represents an attack surface for indirect prompt injection. * Ingestion points: Sub-agent outputs relayed via the 'Subagent-relay rewrite rule' in SKILL.md. * Capability inventory: Access to shell execution (git, curl, gbrain) and file system operations. * Boundary markers: Absent for sub-agent output blocks. * Sanitization: The instructions mandate path rewriting but do not specify content escaping or validation.
Audit Metadata