DailyOps

DailyOps workflows automate incremental progress toward large goals through small, scheduled daily changes. Work happens automatically in manageable pieces that are easy to review and integrate.

• Continuous improvement - Daily code quality improvements
• Progressive migrations - Gradually update dependencies or patterns
• Documentation maintenance - Keep docs fresh with daily updates
• Technical debt - Chip away at issues one small PR at a time

Workflows run on weekday schedules (avoiding weekends) with workflow_dispatch enabled for manual testing:

---
on:
  schedule:
    - cron: "0 2 * * 1-5"  # Weekdays only (no short syntax available)
  workflow_dispatch:
---

Work proceeds through three phases with maintainer approval between each:

1. Research - Analyze state, create discussion with findings
2. Configuration - Define steps, create config PR
3. Execution - Make improvements, verify, create draft PRs

Use GitHub discussions to maintain continuity across runs. The workflow creates a discussion (if none exists) and adds progress comments on subsequent runs:

safe-outputs:
  create-discussion:
    title-prefix: "${{ github.workflow }}"
    category: "ideas"

The safe-outputs: (validated GitHub operations) configuration lets the AI request discussion creation without requiring write permissions.

For workflows that post updates to an existing discussion, use add-comment with discussion: true and a specific target discussion number:

safe-outputs:
  add-comment:
    target: "4750"
    discussion: true

This pattern is ideal for daily status posts, recurring reports, or community updates. The daily-fact.md workflow demonstrates this by posting daily facts about the repository to a pinned discussion thread.

Enable cache-memory to maintain state at /tmp/gh-aw/cache-memory/ across runs, useful for tracking progress, storing metrics, and building knowledge bases over time:

tools:
  cache-memory: true

This repository implements several DailyOps workflows demonstrating different use cases:

• daily-fact.md - Posts daily facts about the repository to a discussion thread
• daily-test-improver.md - Systematically adds tests to improve coverage incrementally
• daily-perf-improver.md - Identifies and implements performance optimizations
• daily-doc-updater.md - Keeps documentation synchronized with merged code changes
• daily-team-status (from agentics) - Creates daily team status reports with activity summaries
• daily-repo-chronicle.md - Produces newspaper-style repository updates
• daily-firewall-report.md - Analyzes and reports on firewall activity

All follow the phased approach with discussions for tracking and draft pull requests for review.

1. Define Goal - Identify ongoing goal (test coverage, performance, docs sync)
2. Design Workflow - Set weekday schedule, configure safe-outputs for discussions/PRs
3. Research Phase - Analyze state, create discussion, wait for approval
4. Config Phase - Create config files, test, submit PR, wait for approval
5. Execute Daily - Make small improvements, verify, create draft PRs, update discussion

• IssueOps - Trigger workflows from issue creation or comments
• ChatOps - Trigger workflows from slash commands in comments
• LabelOps - Trigger workflows when labels change on issues or pull requests
• Planning Workflow - Use /plan command to split large discussions into actionable work items, then assign sub-tasks to Copilot for execution

DailyOps complements these patterns by providing scheduled automation that doesn’t require manual triggers.