fixing-code

Pass

Audited by Gen Agent Trust Hub on Feb 28, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses the Bash tool to run standard linting and testing utilities like make, go, ruff, pytest, and bun. Execution is restricted to these specific command patterns.
  • [PROMPT_INJECTION]: The skill has an indirect prompt injection surface where it passes raw output from external tools into the prompts of subagents.
  • Ingestion points: Phase 1 captures output from shell commands in files and tool outputs.
  • Boundary markers: No explicit delimiters or ignore instructions are used to separate tool output from subagent prompts.
  • Capability inventory: The skill can execute restricted bash commands, spawn tasks, and edit files.
  • Sanitization: No sanitization or filtering of the captured command output is performed before interpolation into agent prompts.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 28, 2026, 04:50 PM