doc-updater

The doc-updater skill is a proactive documentation maintenance tool designed for software developers and technical leads who need to maintain synchronization between evolving codebases and their supporting technical documentation. By analyzing Git diffs and applying pattern-matching heuristics, the agent identifies specific areas of a project—such as CLI interfaces, API endpoints, environment variables, and configuration schemas—that require updates to remain accurate. This tool reduces the cognitive load on developers by automating the tedious process of cross-referencing code modifications with documentation files like README.md, CHANGELOG.md, and dedicated docs/ folders. It is particularly effective for teams following strict version control workflows or pre-PR checklists, ensuring that documentation is not an afterthought but an integral part of the development cycle.

• Automatically scans changed code using regex patterns to identify modified API routes, new CLI options, or updated environment variables.

• Provides structured, Markdown-formatted suggestions using diff-style syntax to make applying changes seamless.

• Integrates with standard repository structures, supporting documentation in Markdown, README files, API specifications, and configuration guides.

• Features a built-in interactive workflow, presenting a 'Documentation Impact Report' that allows users to review and approve specific update suggestions before they are executed.

• Encourages best practices by keeping the CHANGELOG updated for breaking changes, new features, and security patches.

• Facilitates integration with secondary tools like code-quality-gates and automated PR checklist tools.

• Intended for developers working in Python-based environments or projects using standard Git workflows where documentation lives in the repository.

• Requires existing documentation files (e.g., .md files, docs/ directory) to be present for the tool to detect and propose edits.

• Uses standard Git operations like git diff to track changes, so it is best used within initialized repositories.

• While highly effective for structural updates, users should always manually verify complex or conceptual documentation changes that require domain-specific context not present in the code.

• Supports custom detection patterns via Python scripts within the skill configuration to tailor discovery to specific project needs.