The Agent Skills Directory

[SAFE]: The skill provides a rigorous framework for software verification, focusing on intent alignment and evidence-based reporting.
[SAFE]: The included utility scripts/fingerprint.py uses only standard Python libraries (hashlib, json, re, sys, pathlib) to calculate file hashes for integrity verification. It does not perform any network operations, dynamic code execution, or unauthorized file system access.
[SAFE]: The instructions explicitly mandate the use of isolated environments, such as fresh subagents or sessions, and include warnings against running consequential checks in production environments, which aligns with security best practices for automated verification.
[SAFE]: No obfuscation, prompt injection, credential harvesting, or persistence mechanisms were identified in the skill instructions, scripts, or metadata.

sdlc-verify