debug-loop
Pass
Audited by Gen Agent Trust Hub on Sep 18, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructions require the agent to execute shell commands for reproducing bugs, querying databases with sqlite3, and running regression tests with npm. These capabilities are necessary for debugging but provide a broad interface for system interaction.
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process data from external environments, creating a surface for potential injection attacks.
- Ingestion points: The agent captures and displays raw output from reproduction commands in Phase 1 and database query results in Phase 2.
- Boundary markers: Absent. The instructions lack delimiters or explicit directives for the agent to ignore or isolate instructions found within captured external output.
- Capability inventory: The agent is empowered to modify the codebase in Phase 3, search the filesystem using grep, and execute arbitrary commands for verification and regression.
- Sanitization: Absent. The skill captures real output from the system without any sanitization, filtering, or validation of the content before it enters the agent context.
Audit Metadata