The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes local git commands via Python's subprocess module to collect repository context. This is the intended functionality and is implemented using secure practices (passing arguments as a list, avoiding shell=True) to prevent command injection, even when handling user-provided repository paths or references.
[DATA_EXPOSURE]: The scripts analyze commit messages and file paths to identify potential risks (e.g., changes to .env files or database schemas). This information is presented to the user/agent as context for generating release notes and is not exfiltrated to external services.
[SAFE]: No malicious patterns, such as obfuscation, unauthorized network access, prompt injection, or persistence mechanisms, were detected. The skill uses standard Python libraries and operates within the expected scope of a local developer utility.

release-notes