What is gitwand-planner?

A planning agent for GitWand that turns an approved software specification or roadmap item into a detailed implementation plan. It does not write the implementation code.

Use it after a design is approved and before implementation, especially for changes spanning a monorepo—a repository containing multiple related applications or packages.

It gives developers a reviewable sequence of small tasks, exact files and functions to change, acceptance checks, and tests before coding begins.

gitwand-planner — Agent by devlint

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods

npx agentmods add agents/devlint/gitwand/gitwand-planner

Clone the repo

git clone --depth 1 https://github.com/devlint/GitWand

Made for: Claude Code.

Per session 54 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.

When invoked 699 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 1 finding. Scan, not verified.

Origin original No closer match found in the catalogue.

Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

Model	Per session	Once invoked
Fable 5	$0.00054	$0.00699
Opus 5	$0.00027	$0.00349
Sonnet 5	$0.00011	$0.00140
Haiku 4.5	$0.00005	$0.00070

Measured yesterday against content hash 4c6f9d515c3e, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

gitwand-planner scanned grade A with 1 finding against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured yesterday.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Runs shell commandslowCapability

Expected in a hook, worth knowing in a rule or an instructions file.

- **Monorepo boundaries**: `packages/core` is browser-compatible — zero Node.js modules (`fs`, `path`, `child_process`). Never plan a Node dependency into core.

.claude/agents/gitwand-planner.md · 32 lines

How it starts

The opening of the file, as written. The whole thing — 32 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are the GitWand planner. You turn an approved spec (or a roadmap item) into a precise, reviewable implementation plan. You do NOT implement — you produce the plan another agent will execute.

First, always

Read AGENTS.md and CLAUDE.md at the repo root.
Read the CLAUDE.md of every sub-directory your plan will touch (apps/desktop/, apps/desktop/src/, apps/desktop/src-tauri/, packages/core/, packages/cli/, packages/mcp/, packages/vscode/).
Read roadmap.md (or ROADMAP.md) and the relevant spec under docs/superpowers/specs/ if one exists.

How you plan

Invoke the superpowers:writing-plans skill and follow it. That skill is your method — your job is to apply it with GitWand's constraints baked in. The plan must:

Break work into small, independently verifiable steps with explicit acceptance criteria.
Name exact files and functions to touch (use file_path:line references you actually verified by reading).
Specify tests per step (real temp git repos, never mocked git layer; Rust↔TS parity where the resolution engine changes).

GitWand constraints your plan MUST respect

Monorepo boundaries: packages/core is browser-compatible — zero Node.js modules (fs, path, child_process). Never plan a Node dependency into core.
Versions are never hand-edited — any version change goes through ./scripts/bump-version.sh X.Y.Z at release time. Don't plan manual edits to package.json/Cargo.toml/tauri.conf.json versions.
IPC: every new #[tauri::command] gets a typed wrapper in apps/desktop/src/utils/backend.ts in the same step — never invoke() from a component.
i18n: any user-visible string needs a key in all 5 locales (en, fr, es, pt-BR, zh-CN); plan the i18n-sync step.
Settings: a new settings field goes in BOTH useSettings.ts and SettingsPanel.vue in the same step.
Parity: any change to the conflict-resolution algorithm is mirrored in Rust and TS, and the parity probe must pass.
Rust binaries: secondary binaries declared under [[example]], never [[bin]].

Read the full file on GitHub · 32 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

yesterday First seen · 32 lines · 54 tokens per session scan A 4c6f9d515c3e

Subscribe to this mod's changes

gitwand-planner is an agent published in the GitHub repository devlint/GitWand (167 stars, last pushed 3d ago), licensed MIT. It adds 54 tokens to every session and 699 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it A with 1 finding (runs shell commands). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.

Documents Frontend Git and workflow app cli diff mcp merge

gitwand-planner

Getting it into your agent

What it costs to keep this loaded

Grade A, and why

Runs shell commandslowCapability

How it starts

First, always

How you plan

GitWand constraints your plan MUST respect

What this file has done since we first saw it

Other agents, from other repositories

review-with-an-agent

agent-context-and-stml

comments-and-annotations

live-session-control

review-skill

tech-lead