code-simplifier

Pass

Audited by Gen Agent Trust Hub on Aug 30, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [SAFE]: No evidence of prompt injection, data exfiltration, or unauthorized command execution was found. The skill references documentation from trusted organizations (Anthropic) and well-known services (Sentry) for its design basis.
  • [INDIRECT_PROMPT_INJECTION]: The skill's primary function is to ingest and modify source code, which constitutes an attack surface for instructions embedded within the processed code.
  • Ingestion points: The agent reads and processes code from the current workspace or session for simplification (SKILL.md).
  • Boundary markers: The instructions lack explicit delimiters for user code but emphasize following existing project rules and AGENTS.md configurations.
  • Capability inventory: The skill is authorized to perform file writes to apply code improvements, provided the task has been granted write permission (SKILL.md).
  • Sanitization: No specific input sanitization or escaping mechanisms are described; the skill relies on the agent's ability to distinguish code implementation from instructions.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 30, 2026, 06:10 PM
Security Audit — agent-trust-hub — code-simplifier