spring-cleaning

Pass

Audited by Gen Agent Trust Hub on Apr 13, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill is vulnerable to Indirect Prompt Injection as it processes a codebase to identify and delete 'dead' code and files. Malicious instructions could be embedded in the codebase to manipulate the agent's cleanup logic.
  • Ingestion points: The agent reads the local codebase, including all source files and configuration.
  • Boundary markers: Absent. The instructions do not provide delimiters or warnings to ignore commands within the codebase.
  • Capability inventory: Includes file deletion and structural code modification across the project.
  • Sanitization: Absent. There is no validation or filtering of the content being analyzed before it is used to justify deletions or refactoring.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 13, 2026, 04:16 PM
Security Audit — agent-trust-hub — spring-cleaning