craft-defaults

Pass

Audited by Gen Agent Trust Hub on Sep 20, 2026

Risk Level: SAFENO_CODE
Full Analysis
  • [SAFE]: The skill consists entirely of natural language instructions regarding engineering best practices and stylistic choices, such as prioritizing modularity and maintainability over implementation time.
  • [NO_CODE]: No scripts, executables, or code snippets are provided within the skill. It defines a "posture" for the agent rather than providing tools or automated scripts.
  • [PROMPT_INJECTION]: No attempts to bypass safety filters, extract system prompts, or override core agent guidelines were detected. The instructions are focused on software development hygiene (e.g., bug reproduction end-to-end, avoiding manual edits to auto-generated files).
  • [DATA_EXFILTRATION]: The skill does not contain any network operations, references to sensitive files, or instructions to transmit data externally.
  • [DYNAMIC_CONTEXT_INJECTION]: No dynamic context placeholders or command execution patterns were found in the markdown content.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 20, 2026, 08:52 PM
Security Audit — agent-trust-hub — craft-defaults