GitHub Agentic Workflows

Automated dependency analysis

Go Fan is a weekday dependency-analysis workflow for Go repositories. On each run, it selects a direct dependency from go.mod, researches upstream changes and recommended usage, compares those findings with the repository’s code, and creates a single issue containing actionable recommendations.

flowchart LR
    A[Select direct Go module] --> B[Research upstream guidance]
    B --> C[Analyze repository usage]
    C --> D[Create follow-up issue]

Go Fan tracks previously reviewed modules with cache-memory. It prioritizes recently updated GitHub-hosted dependencies, then cycles through the remaining direct dependencies. It avoids selecting a module reviewed in the last seven days unless every direct dependency is still within that review window, in which case it resets the list and starts from the top.

The workflow saves a detailed module summary under scratchpad/mods/ and creates a [go-fan] issue. The issue identifies current usage, relevant upstream changes, best-practice gaps, and prioritized opportunities for simplification, improved error handling, configuration, testing, or performance.

Review the recommendations before implementing them. For a structured handoff from research to planning, assignment, and human review, use the ResearchPlanAssignOps pattern.

The active workflow runs on weekdays and can also be started manually: