For an overview of all available workflows, see the main README.
The Glossary Maintainer workflow automatically maintains project glossary or terminology documentation by scanning code changes and keeping technical terms up-to-date.
# Install the 'gh aw' extension
gh extension install github/gh-aw
# Add the workflow to your repository
gh aw add-wizard githubnext/agentics/glossary-maintainerThis walks you through adding the workflow to your repository.
You can start a run of this workflow immediately by running:
gh aw run glossary-maintainerThe Glossary Maintainer workflow runs daily on weekdays and:
- Scans Recent Changes - Reviews commits and PRs from the last 24 hours (daily) or 7 days (weekly on Mondays)
- Identifies New Terms - Finds technical terminology, configuration options, and project-specific concepts
- Updates Definitions - Adds new terms or updates existing definitions based on code changes
- Maintains Consistency - Ensures glossary follows existing structure and style
- Creates Pull Requests - Proposes glossary updates for review when new terms are found
- Daily (Mon-Fri): Incremental scan of last 24 hours
- Monday: Full scan of last 7 days for comprehensive review
The workflow identifies terms that are:
- Used in user-facing documentation or code
- Project-specific or domain-specific
- Technical terms requiring explanation
- Newly introduced in recent changes
The workflow avoids adding:
- Generic programming terms
- Self-evident terminology
- Internal implementation details
- Terms only in code comments
This workflow works out of the box and automatically:
- Locates your glossary file (common paths:
docs/glossary.md,docs/reference/glossary.md,GLOSSARY.md) - Follows your existing glossary structure and style
- Maintains alphabetical or categorical organization
- Uses cache memory to avoid duplicate work
You can edit the workflow to customize:
- The glossary file path if it's in a non-standard location
- The scanning timeframe (daily vs weekly)
- The PR title prefix and labels
After editing, run gh aw compile to apply your changes.
- Recent commits and their diffs
- Merged pull requests from the specified timeframe
- PR descriptions and comments
- Code changes that introduce new terminology
- Pull requests with glossary updates
- Cache memory of processed commits to avoid duplicates
- Daily on weekdays (incremental scan of last 24 hours)
- Mondays (full scan of last 7 days)
- Manually via workflow_dispatch
contents: read- To read repository filesissues: read- To review issue discussionspull-requests: read- To analyze PR descriptionsactions: read- To check workflow runs
- Automatic maintenance: Keeps terminology documentation current without manual tracking
- Consistency: Ensures definitions stay aligned with code changes
- Accessibility: Helps new contributors understand project-specific terms
- Time savings: Eliminates manual glossary updates
- Historical context: Uses cache memory to track what's been processed
When the workflow finds new terms, it creates a PR like:
Title: [docs] Update glossary - daily scan
Body:
### Glossary Updates
**Scan Type**: Incremental (daily)
**Terms Added**:
- **Safe Outputs**: Security mechanism for controlling workflow write operations
- **Cache Memory**: Persistent storage for workflow state across runs
**Terms Updated**:
- **Frontmatter**: Updated to include new engine configuration options
**Changes Analyzed**:
- Reviewed 5 commits from last 24 hours
- Analyzed 2 merged PRs
- Processed 1 new feature (cache-memory tool)
**Related Changes**:
- abc1234: Add cache-memory tool support
- PR #123: Implement safe-outputs validationThis workflow is adapted from Peli's Agent Factory. Read more: Meet the Workflows: Continuous Documentation
This workflow is valuable for:
- Open source projects with growing terminology
- Technical documentation that needs to stay synchronized with code
- API projects with evolving configuration options
- Developer tools with CLI commands and options
- Frameworks with specialized concepts and patterns
Possible improvements:
- Suggest related terms when adding new definitions
- Detect inconsistent terminology usage across documentation
- Generate glossary if one doesn't exist
- Link term usage to specific code locations
- Track term popularity and usage frequency