skills/pietz/skills/simplify-codebase/Gen Agent Trust Hub

simplify-codebase

Pass

Audited by Gen Agent Trust Hub on Jul 10, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill instructs the agent to ingest and analyze existing source code and tests, which could contain malicious instructions designed to influence the agent's behavior during the refactoring process.
  • Ingestion points: Workflow steps 2, 3, and 5 in SKILL.md require the agent to read existing codebase modules and tests.
  • Boundary markers: The instructions do not define specific delimiters or instructions to ignore embedded prompts within the source code being analyzed.
  • Capability inventory: The skill involves codebase analysis, architectural design, and staged code implementation, which assumes the agent has file-system read and write capabilities.
  • Sanitization: The workflow implements a form of 'clean room' design in Step 4, where the design phase is performed by an isolated agent given only a behavioral contract and no direct access to the original code, significantly mitigating the risk of instructions leaking from the old code into the new design.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 10, 2026, 02:12 PM
Security Audit — agent-trust-hub — simplify-codebase