ts-workflow-explore-plan-review
Pass
Audited by Gen Agent Trust Hub on Aug 4, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [SAFE]: The skill implements a robust router pattern that emphasizes human oversight via a design gate. It does not request excessive privileges, access sensitive environment secrets, or perform unauthorized network operations.
- [PROMPT_INJECTION]: The skill has an indirect prompt injection surface because it relays user input to internal sub-agents. \n- Ingestion points: User input enters the system through the original request and subsequent feedback during the relay loop in SKILL.md. \n- Boundary markers: The skill does not use specific delimiters or instructions to ignore embedded commands when relaying content to the exploration judge. \n- Capability inventory: The skill dispatches the ts-explore sub-agent and writes design artifacts to the local file system (docs/designs/). \n- Sanitization: There is no explicit sanitization or filtering logic applied to the user input before it is interpolated into the agent prompts.
Audit Metadata