bugsweep
Pass
Audited by Gen Agent Trust Hub on Nov 13, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [SAFE]: No malicious patterns or security risks were identified in the skill's logic or metadata. The author provides a legitimate quality-assurance workflow.
- [INDIRECT_PROMPT_INJECTION]: The skill ingests external project code (from directories like src/ and lib/) to identify bugs. This presents an inherent attack surface for indirect prompt injection. Evidence Chain: 1. Ingestion points: Project source files (src/, lib/). 2. Boundary markers: None explicitly defined. 3. Capability inventory: Git operations, file system writes (.bak), and code execution (pytest, smoke runs). 4. Sanitization: None. The protocol's requirement for verification mitigates the risk of the agent following malicious embedded instructions.
- [DYNAMIC_EXECUTION]: The workflow requires the agent to execute code via pytest and manual smoke runs to verify bug fixes. This is the intended primary purpose of the skill and is performed within the context of the user's development project.
Audit Metadata