source-command-fix-build
Pass
Audited by Gen Agent Trust Hub on Jun 23, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill implements a standard software development workflow for fixing build errors. It includes clear constraints to ensure the agent only performs the minimum necessary changes to resolve a specific issue, which helps prevent unintended code modifications.
- [PROMPT_INJECTION]: The skill processes untrusted data from build command outputs (stderr/stdout) to diagnose errors. While this constitutes an attack surface for indirect prompt injection, it is essential for the tool's primary purpose and no specific malicious intent was found.
- Ingestion points: Full command failure output is ingested from the environment and passed to a sub-agent (defined in SKILL.md).
- Boundary markers: Absent; the instructions do not specify the use of delimiters to isolate the command output from the sub-agent's instructions.
- Capability inventory: The sub-agent (build-error-resolver) is authorized to analyze project configuration files and apply code diffs to fix identified errors.
- Sanitization: No validation or sanitization of the command output is performed before it is analyzed by the agent.
Audit Metadata