The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes standard git commands (git rev-parse, git diff, git diff --cached) to locate the repository root and identify code changes for documentation.
[PROMPT_INJECTION]: The skill reads external data via git diff output to determine changelog entries. While this creates a potential surface for indirect prompt injection from malicious code comments or strings in the diff, the impact is limited to the content of the CHANGELOG.md file. 1. Ingestion points: Step 3 reads output from git diff and git diff --cached in SKILL.md. 2. Boundary markers: No specific delimiters are used to separate diff content from the agent's instructions. 3. Capability inventory: The skill can read repository files, execute git status/diff commands, and write to CHANGELOG.md. 4. Sanitization: No explicit sanitization of the diff content is performed before processing.

update-changelog