The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from the repository's git log and various configuration files, creating a potential surface for indirect prompt injection attacks.
Ingestion points: The skill reads external data from git log --oneline and multiple project files such as package.json, Cargo.toml, and others discovered during the project structure analysis.
Boundary markers: The instructions do not specify any delimiters or safety markers to help the agent distinguish between its own instructions and content found within the analyzed logs or files.
Capability inventory: The skill is granted significant capabilities including the ability to Read, Write, and Edit files, as well as executing git commands like git add and git commit.
Sanitization: There is no requirement for the agent to sanitize or filter content from commit messages before they are processed for versioning logic or written to the CHANGELOG.md file.

bump-version