By primus852
Generate end-user manuals (Markdown) with screenshots and flowcharts from Flutter codebases.
Route a freeform documentation request to the right sub-workflow. Use when the user asks something like "add a chapter for the login screen", "re-capture screenshots", "regenerate diagrams", "update the PDF", or any other partial-pipeline request. Never does the work itself — always delegates.
Generate an end-user manual for the current Flutter project. Use this command when the user says "generate docs", "create a manual", "document this project", or "write a user guide". Runs the full pipeline: code analysis → screenshots → diagrams → prose drafting. Output is a folder of cross-linked Markdown files.
Show full reference for the flutter-doc plugin — all commands, options, output formats, audience flags, and config keys. Use when the user asks "how do I use flutter-doc", "what options does /flutter-doc:document have", or "documenter help".
Check for flutter-doc updates, show changelog, and install if the user confirms. Use when the user says "update flutter-doc", "check for updates", or "/flutter-doc:update".
Analyzes a Flutter codebase to extract the user-facing surface: routes, screens, forms, buttons, and multi-step user journeys. Works with MaterialApp routes and GoRouter config. Returns two JSON files: routes.json and journeys.json. Spawn this agent at the start of any documentation pipeline — it produces the structured data that all downstream agents consume.
Converts journeys.json into Mermaid flowchart and sequenceDiagram blocks, then renders them to SVG files using the mmdc CLI. One diagram per user journey. Spawn this agent after code-cartographer — it reads journeys.json and writes SVG files to .documenter/diagrams/.
Writes the end-user manual as a folder of cross-linked Markdown files from structured analysis data (routes.json, journeys.json, screenshots, diagrams). Produces index.md, one chapter file per feature group, one journey file per how-to walkthrough, plus troubleshooting.md and getting-help.md. Spawn this agent after code-cartographer, screenshot-orchestrator, and diagram-designer have run. The drafter loads the manual-authoring skill for style rules.
Captures screenshots of each Flutter screen in the project. Invokes `flutter screenshot` on a connected device/emulator for each route via deep-link navigation, with a graceful fallback to a user-provided manual screenshot folder. Writes a manifest.yaml inventory. Spawn this agent after code-cartographer.
Procedural knowledge for analyzing Flutter app codebases to extract user-facing screens, routes, form fields, and navigation. Use this skill when the code-cartographer agent needs guidance on Flutter-specific extraction patterns. Covers MaterialApp routes, GoRouter config, and widget text extraction. Skip for web or backend codebases.
Procedural knowledge for writing end-user manuals. Use this skill when writing, editing, or reviewing user-facing documentation chapters, getting-started guides, how-to sections, or troubleshooting entries. Covers tone, voice, chapter structure, audience calibration, and screenshot/diagram integration. Skip when the user wants API reference docs, code comments, or developer-facing documentation.
Uses power tools
Uses Bash, Write, or Edit tools
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
A Claude Code plugin that reads your Flutter codebase and produces a cross-linked end-user manual as a folder of Markdown files — complete with screenshots and Mermaid flowcharts.
Writing user documentation for a Flutter app is tedious: screens multiply, flows change, and prose goes stale. claude-flutter-documenter automates the first draft by letting an LLM do what it's good at — reading code, drafting prose, and keeping structure consistent — while you stay in charge of accuracy.
One command turns a Flutter project into a ready-to-review manual:
MaterialApp / GoRouter analysisflutter screenshot on a connected device or emulatormmdc is installedAll output lands in .documenter/ inside your project and opens without any build step in VS Code, GitHub, or Obsidian.
| Requirement | Version | Notes |
|---|---|---|
| Claude Code | Latest | Plugin host |
| Node.js | ≥ 18 | Runs the analysis CLI |
| Flutter SDK | Any stable | Must be in PATH |
Mermaid CLI (mmdc) | Any | Optional — enables SVG diagram export |
/plugin marketplace add https://github.com/primus852/claude-flutter-documenter
/plugin install flutter-doc
brew install mermaid-cli # enables SVG diagram export (optional)
For local development — clone the repo and point the marketplace at your checkout instead:
git clone https://github.com/primus852/claude-flutter-documenter ~/claude-flutter-documenter /plugin marketplace add ~/claude-flutter-documenter /plugin install claude-flutter-documenter
Open a Flutter project in Claude Code (the working directory must contain pubspec.yaml), then run:
/document
The plugin walks through six steps automatically:
pubspec.yaml).documenter/ output structure/flutter-doc:document # guided run, auto-detects Flutter project
/flutter-doc:document end-user # audience = end-user (default)
/flutter-doc:document admin # audience = admin
/flutter-doc:document power-user # audience = power-user
/flutter-doc:document --src lib/features/auth # scope analysis to a subfolder
/flutter-doc:help # full command reference
/flutter-doc:do "update the settings chapter" # freeform refinement
| Audience | Description |
|---|---|
end-user (default) | Step-by-step, plain language, screenshot on every screen |
admin | Technical detail, assumes familiarity with the system |
power-user | Keyboard shortcuts and advanced flows |
--src flagScope analysis to a specific subfolder when your app is large or monorepo-structured:
/document --src lib/features/onboarding
The path is saved in .documenter/config.json and becomes the default for all future runs on that project.
The sample/ directory contains a complete documenter run against the Flutter counter demo app (flutter create):
sample/manual/index.md — generated manual entry pointsample/diagrams/overview.svg — Mermaid flowchartsample/analysis/routes.json — extracted routessample/screenshots/home.png — captured screenshot.documenter/
├── config.json ← project settings (audience, locale, …)
├── analysis/
│ ├── routes.json ← discovered routes / screens
│ ├── journeys.json ← multi-step user flows
│ └── manifest.yaml ← screenshot inventory
├── screenshots/
│ └── <route-id>.png
├── diagrams/
│ ├── <journey-id>.mmd ← Mermaid source (always written)
│ └── <journey-id>.svg ← only if mmdc is installed
└── manual/
├── index.md ← start here
├── chapters/
│ └── <chapter-id>.md
├── journeys/
│ └── <journey-id>.md
├── troubleshooting.md
└── getting-help.md
Open .documenter/manual/index.md in any Markdown viewer — all internal links resolve without a build step.
npx claudepluginhub primus852/claude-flutter-documenter --plugin flutter-docClaude + Obsidian knowledge companion. Sets up a persistent, compounding wiki vault (Karpathy's LLM Wiki pattern). v1.7 "Compound Vault" + v1.8 methodology modes close 5 of 5 priority gaps from the May 2026 compass artifact. Ships: substrate alignment with kepano/obsidian-skills, default Obsidian CLI transport, hybrid retrieval (contextual prefix + BM25 + cosine rerank per Anthropic's Sept 2024 research), per-file advisory locking for multi-writer safety, pre-commit verifier agent, AND methodology modes (LYT / PARA / Zettelkasten / Generic) for first-class organizational support no other Claude+Obsidian competitor offers. v1.7.x audit closure: every BLOCKER + HIGH + MEDIUM + LOW finding from the v1.7.0 audit is CLOSED or DEFERRED-with-rationale. Optional DragonScale Memory extension (log folds, deterministic addresses, semantic tiling lint, boundary-first autoresearch).
Make your AI agent code with your project's architecture, rules, and decisions.
Upstash Context7 MCP server for up-to-date documentation lookup. Pull version-specific documentation and code examples directly from source repositories into your LLM context.
AI-powered wiki generator for code repositories. Generates comprehensive, Mermaid-rich documentation with dark-mode VitePress sites, onboarding guides, deep research, and source citations. Inspired by OpenDeepWiki and deepwiki-open.
Complete AI coding workflow system. Self-correcting memory + persistent FTS5-indexed research wikis + auto-research loop + multi-LLM council on a single SQLite store. 33 skills, 8 agents, 22 commands, 37 hook scripts across 24 events. Cross-agent via SkillKit.
Comprehensive C4 architecture documentation workflow with bottom-up code analysis, component synthesis, container mapping, and context diagram generation