gh-tasks
gh-tasks
Overview
Issue-driven development pipeline for this monorepo with three modes of operation:
Daily edits - Small changes happen directly on main
Focused work - Features and fixes use issue-linked worktrees with draft PRs:
new → plan → work → review → merge
│ │
└──────────────────────────────┘
(via status dashboard)
Automated pipeline - End-to-end execution via orchestrate sub-skill
Workflow paths
| Path | When to use | Sub-skills |
|---|---|---|
| Quick | Simple bug fix | new → work → merge |
| Standard | Most features | new → work → review → merge |
| Full | Complex work | new → plan → work → review → merge |
| Automated | Well-defined tasks | orchestrate (handles all stages) |
Key principles
- Airtight tracking: one issue = one worktree = one draft PR = one branch
- Status visibility: dashboard shows all active work at a glance
Sub-skills
CRITICAL: You MUST load the appropriate sub-skill from the sub-skills/ directory based on user intent.
Pipeline (in order)
-
new.md: Create a GitHub issue for focused work
- Triggers: "new issue", "create issue", "new task"
-
plan.md: Research codebase and create implementation plan
- Triggers: "plan", "research", "design approach", "analyze options"
-
work.md: Create worktree, implement changes, create draft PR
- Triggers: "work on", "implement", "start coding", "create worktree"
-
review.md: Review PR for correctness, post findings
- Triggers: "review", "check PR", "analyze changes"
-
merge.md: Merge PR, cleanup worktree/branch, close issue
- Triggers: "merge", "finish", "done", "complete"
-
rebase.md: Squash-rebase feature branch onto main
- Triggers: "rebase", "make mergeable", "update branch", "sync with main"
Utilities
-
setup.md: Bootstrap repo labels and configuration
- Triggers: "setup labels", "create labels", "bootstrap repo", "configure labels"
-
tasks-status.md: Dashboard of worktrees, issues, PRs, and next actions
- Triggers: "status", "dashboard", "show active work", "what am I working on"
-
orchestrate.md: Automated end-to-end pipeline
- Triggers: "orchestrate", "auto", "end-to-end", "fully automated"
-
worktree-sync.md: Sync worktrees from remote branches across machines
- Triggers: "sync worktrees", "sync all worktrees", "update worktrees", "fetch all branches"
Process
- Determine user intent from their request
- Load the appropriate sub-skill from
sub-skills/ - Execute the sub-skill process
- Verify expected outcome was achieved
Resources
- sub-skills/: Workflow-specific instructions
- scripts/: Standalone CLI tools matching each sub-skill
- templates/: Output format templates for issues, plans, and reviews
Guidelines
- Airtight tracking: one issue = one worktree = one draft PR = one branch
- Always get user approval before destructive operations
Appendix
Scripts
| Script | Sub-skill | Usage |
|---|---|---|
setup-labels.sh |
setup | ./setup-labels.sh [--dry-run|--delete] |
work.sh |
work | ./work.sh <issue-number> |
merge.sh |
merge | ./merge.sh <pr-number> |
rebase.sh |
rebase | ./rebase.sh [worktree-path] |
tasks-status.sh |
tasks-status | ./tasks-status.sh |
monorepo-worktree-sync |
worktree sync | monorepo-worktree-sync (auto-runs every 2 min) |
Templates
| Template | Purpose |
|---|---|
issue.md |
Goal/Context/Acceptance criteria format for issues |
plan.md |
Implementation plan format (appended to issue body) |
review.md |
PR review comment format with blockers/nice-to-haves |
More from nikhilmaddirala/gtd-cc
tools-catppuccin
Agent skill for creating and validating Catppuccin theme ports
18obsidian-gtd
Obsidian vault management and GTD workflows. Use when integrating with Obsidian vaults, managing notes, organizing knowledge, or supporting Getting Things Done methodology through note-based workflows.
13web-search
General web search patterns and techniques including Gemini CLI coordination. Use this skill when you need to perform web searches, find current information, or research topics online. Covers both Gemini CLI and built-in WebSearch tool usage with precise instruction crafting.
11tools-diagnostics
Interactive system resource analysis and troubleshooting for memory, disk, CPU, and performance issues
11web-content-extraction
Extract documentation and content from websites. Supports Mintlify, Starlight/Astro, Docusaurus, GitBook, ReadTheDocs, Sphinx, and generic sites. Uses a tiered approach - try the simplest method first (direct curl, Jina AI Reader) before falling back to Crawl4AI for JS-heavy sites.
10docs-pdf
Parse PDF documents into repository-friendly markdown and text artifacts. Use when users need to extract text, tables, or structure from PDF files.
10