The Agent Skills Directory

[Indirect Prompt Injection] (HIGH): The skill processes git commit logs, which are external data that can be influenced by third parties.
Ingestion points: Git commit messages (via git log) and local configuration files like CHANGELOG_STYLE.md.
Boundary markers: None identified; there are no instructions provided to the agent to treat commit data as untrusted or to use delimiters.
Capability inventory: The skill likely requires shell access to run git commands and filesystem access to read project files.
Sanitization: No evidence of sanitization or filtering of commit messages is present.
[Command Execution] (MEDIUM): To 'Scan Git History', the skill must execute system-level commands. If the agent accepts unsanitized user input for version tags or branch names, it could be vulnerable to command injection.

changelog-generator