The Agent Skills Directory

[SAFE]: The skill uses standard command-line tools such as git, git grep, and ast-grep to perform local repository analysis. These operations are scoped to documentation maintenance and do not involve unauthorized file access or network communication.
[SAFE]: A 'safe-fix' boundary is explicitly defined, restricting the agent to only perform mechanical text substitutions for version numbers and changelog updates. All other documentation drift is flagged for human review, preventing automated injection or unauthorized prose modification.
[SAFE]: No evidence of prompt injection, data exfiltration, or obfuscation was found. The skill ignores sensitive directories like .git and node_modules by default, following security best practices for repository-wide scanning.

sync-docs