Hive Related Work

Hive Related Work

Use this skill to gather prior art for a Hive action from Hive and local git history.

Input

Accept a Hive action URL or raw action ID. For URLs, extract actionId from the query string first, then fall back to any recognizable path segment.

Workflow

1. Resolve the source action.

   • Resolve workspace from user context or Hive MCP.
   • Fetch the action with getActions using specificIds.
   • Capture title, description, labels, project, assignees, and attached branch names.

2. Build high-value search terms.

   • Prefer specific nouns, product names, feature names, API names, config keys, and acronyms.
   • Avoid generic terms such as bug, fix, update, data, error, api, and ui.

3. Search Hive for related actions.

   • Use targeted getActions text searches.
   • Prefer candidates sharing project, labels, branch names, or distinctive terms.
   • Exclude the source action.

4. Search local git history when in a repo.

   • Inspect git diff and git diff --cached for local related changes.
   • Extract stable identifiers: functions, classes, types, constants, config keys, and domain-specific variable names.
   • Search commit messages with git log --all --oneline --grep="<keyword>".
   • Search code history with git log -S"<token>" --all --oneline.
   • Inspect strongest hits with git show <hash> --stat, and patches only when needed.

5. Link commits back to Hive actions when supported.

   • Check branches containing promising commits.
   • Compare branch names against githubBranchNames on source and related actions.
   • Do not claim a commit is related unless message, token history, or patch context supports it.

Output

## Related Work For [Action title]

### Source Action
- [Action title] - [status]
- Key terms: [keyword 1], [keyword 2], [keyword 3]

### Related Hive Actions
- [Action title] - [status] - [why it is related]

### Related Git Commits
- `[hash]` [commit subject] - [why it matters]

### Commit To Action Links
- `[hash]` -> `[branch-name]` -> [Hive action title]

Prefer a small set of strong matches over a long weak list.