The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes local git and shell commands to analyze the repository state.
Evidence: Commands include git log, git tag, git remote get-url origin, cat, and grep.
Context: These operations are necessary for the skill's primary purpose of parsing history and configuration to generate documentation.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from git commit history and existing files which could contain malicious instructions.
Ingestion points: Reads commit messages via git log, existing changelog files (CHANGELOG.md, HISTORY.md), and project configuration files.
Boundary markers: Absent. The instructions do not specify using delimiters or provide warnings to ignore embedded instructions within the processed data.
Capability inventory: The skill can read local files, execute git commands, and write/overwrite the CHANGELOG.md file.
Sanitization: Absent. There are no explicit instructions for the agent to sanitize, escape, or filter content from commit messages or existing files before summarizing them.

changelog-generator