The Agent Skills Directory

[SAFE]: The skill uses local shell commands (cat, sha256sum, test) to manage project metadata and file integrity. These operations are scoped to non-sensitive project directories such as docs/ and .claude/rules/.
[SAFE]: Interactive prompts are implemented via the AskUserQuestion tool for all critical actions, including file regeneration and manifest updates, ensuring the user maintains control over file system changes.
[SAFE]: No network operations, external downloads, or attempts to access sensitive system credentials were found. The skill does not exhibit any patterns of obfuscation or prompt injection.

blueprint-sync