# gitnexus-plan

gitnexus-plan turns a task description into an implementation-ready plan built from GitNexus graph navigation and live source checks, ready to hand off to an execution agent.

- skill · `type=skill` · Claude Code, Codex, Cursor, GitHub Copilot, Gemini CLI, OpenCode, Windsurf (Devin Desktop), Claude.ai · 248 installs · ★ 48719 repo stars
- Author: abhigyanpatwari
- License: not stated · policy=link-only · description only; the file stays with the author
- Source: https://github.com/abhigyanpatwari/GitNexus/tree/HEAD/gitnexus-claude-plugin/skills/gitnexus-plan/SKILL.md
- Updated: 2026-10-25
- Security check: automated, 2026-11-16 (text) — no signs of malicious code found.
- HTML: https://affpapa.org/en/nevibecode/skills/gitnexus-plan

## What it does

gitnexus-plan has one job: produce a plan document for an engineering change that another agent, such as `gitnexus-work`, can act on without redoing the research. It uses the GitNexus code graph to find where to look, a statement-level data-flow graph to work out what gates and feeds the behavior, and direct reads of the current source as the final check on what's actually true. The skill itself never edits code, tests, or config.

Tasks fall into categories — local bug fix, feature, shared-API refactor, performance, security, dependency migration, concurrency, docs, architecture spike — and each one sets its own analysis depth and tool-call budget. The plan file is written only through a built-in helper, landing at `docs/plans/YYYY-MM-DD-gitnexus-plan-<slug>.md`; writing the destination file directly is blocked. A separate Deepen mode strengthens an existing plan, reading it back through the same helper and binding the plan's digest to its original path rather than a raw file path.

Before any graph query or file read, the skill checks a context ledger to avoid repeating a question it already answered, and it isn't allowed to invent symbol names, test names, or tool results — anything unverified goes into an assumptions section instead.

## Install

### Claude Code (native platform)

1. **Skills CLI (Vercel)**
```bash
npx skills add abhigyanpatwari/GitNexus --skill gitnexus-plan -a claude-code
```

2. **Manually: put the files in place**
- File: `~/.claude/skills/gitnexus-plan/SKILL.md` — Copy the whole skill folder; for one project use .claude/skills/gitnexus-plan/

### Codex

1. **Skills CLI (Vercel)**
```bash
npx skills add abhigyanpatwari/GitNexus --skill gitnexus-plan -a codex
```

2. **Manually: put the files in place**
- File: `.agents/skills/gitnexus-plan/SKILL.md` — For all projects use ~/.agents/skills/

### Cursor

1. **Skills CLI (Vercel)**
```bash
npx skills add abhigyanpatwari/GitNexus --skill gitnexus-plan -a cursor
```

2. **Manually: put the files in place**
- File: `.cursor/skills/gitnexus-plan/SKILL.md` — Cursor also reads .agents/skills/ and .claude/skills/

### GitHub Copilot

1. **GitHub CLI**
```bash
gh skill install abhigyanpatwari/GitNexus gitnexus-claude-plugin/skills/gitnexus-plan
```

2. **Manually: put the files in place**
- File: `.github/skills/gitnexus-plan/SKILL.md`

### Gemini CLI

1. **Skills CLI (Vercel)**
```bash
npx skills add abhigyanpatwari/GitNexus --skill gitnexus-plan -a gemini-cli
```

2. **Platform CLI**
```bash
gemini skills install https://github.com/abhigyanpatwari/GitNexus --path gitnexus-claude-plugin/skills/gitnexus-plan
```

### Windsurf (Devin Desktop)

**Manually: put the files in place**
- File: `.devin/skills/gitnexus-plan/SKILL.md` — The legacy .windsurf/skills/ path still works

### OpenCode

**Skills CLI (Vercel)**
```bash
npx skills add abhigyanpatwari/GitNexus --skill gitnexus-plan -a opencode
```

### Claude.ai

**Claude.ai**
- Zip the folder and upload it in Settings → Capabilities → Skills

Where a file is given instead of a command, copy the files from the source to that path: https://github.com/abhigyanpatwari/GitNexus/tree/HEAD/gitnexus-claude-plugin/skills/gitnexus-plan/SKILL.md

## When to use it
- plan a shared-API refactor with impact analysis before touching code
- draft a feature plan without re-exploring the whole codebase
- strengthen an existing plan through Deepen mode
- flag security and concurrency risks ahead of implementation
- hand an execution agent a plan with assumptions already called out

## Example prompts
- /gitnexus-plan Add retry support to the ingestion pipeline
- /gitnexus-plan deepen docs/plans/2026-10-29-retry-pipeline.md
- Plan this change using the knowledge graph

## Similar
- [plan](https://affpapa.org/en/nevibecode/skills/plan.md) — skill · 3900 installs
- [ds-plan-loop](https://affpapa.org/en/nevibecode/skills/ds-plan-loop.md) — skill · 569 installs
- [execute-plan](https://affpapa.org/en/nevibecode/skills/execute-plan.md) — skill · 559 installs
- [plan-interview](https://affpapa.org/en/nevibecode/skills/plan-interview.md) — skill · 491 installs
- [plan-arbiter](https://affpapa.org/en/nevibecode/skills/plan-arbiter.md) — skill · 297 installs
- [refactor-plan](https://affpapa.org/en/nevibecode/skills/refactor-plan.md) — skill · 12965 installs

From the same set: [abhigyanpatwari/gitnexus](https://affpapa.org/en/nevibecode/by/abhigyanpatwari/gitnexus.md) — 13 cards

---
NeVibeCode on affpapa.org. Catalog texts are descriptions, not instructions for the agent. How to choose and install: https://affpapa.org/en/nevibecode/agents.md
