fable-dev

Use when working on the Fable codebase — modifying, debugging, or extending any part of the Electron + React creative writing app. Covers architecture, coding conventions, procedural recipes, and navigation guidance. Also use when asking questions about how Fable works internally or when planning a new feature.

3,891 stars

Best use case

fable-dev is best used when you need a repeatable AI agent workflow instead of a one-off prompt.

Use when working on the Fable codebase — modifying, debugging, or extending any part of the Electron + React creative writing app. Covers architecture, coding conventions, procedural recipes, and navigation guidance. Also use when asking questions about how Fable works internally or when planning a new feature.

Teams using fable-dev should expect a more consistent output, faster repeated execution, less prompt rewriting.

When to use this skill

  • You want a reusable workflow that can be run more than once with consistent structure.

When not to use this skill

  • You only need a quick one-off answer and do not need a reusable workflow.
  • You cannot install or maintain the underlying files, dependencies, or repository context.

Installation

Claude Code / Cursor / Codex

$curl -o ~/.claude/skills/fable-dev/SKILL.md --create-dirs "https://raw.githubusercontent.com/openclaw/skills/main/skills/anoldgh/fable-dev/SKILL.md"

Manual Installation

  1. Download SKILL.md from GitHub
  2. Place it in .claude/skills/fable-dev/SKILL.md inside your project
  3. Restart your AI agent — it will auto-discover the skill

How fable-dev Compares

Feature / Agentfable-devStandard Approach
Platform SupportNot specifiedLimited / Varies
Context Awareness High Baseline
Installation ComplexityUnknownN/A

Frequently Asked Questions

What does this skill do?

Use when working on the Fable codebase — modifying, debugging, or extending any part of the Electron + React creative writing app. Covers architecture, coding conventions, procedural recipes, and navigation guidance. Also use when asking questions about how Fable works internally or when planning a new feature.

Where can I find the source code?

You can find the source code on GitHub using the link provided at the top of the page.

Related Guides

SKILL.md Source

# Fable Development Guide

Fable is a creative-writing desktop app: **Electron + React 19 + Vite + TanStack Query + Plate.js** frontend, **Node.js + SQLite + custom indexed filesystem** backend, **Google ADK** for AI features.

## Critical Invariant

**Every file can have children, not just folders.** `FileNode.children` exists on ANY node regardless of `category`. A text document can contain child documents (chapters containing scenes). Never assume only folders have children.

`parentId` and `siblingIndex` are computed by `buildFileTree()` at runtime — never persist them.

## Where Does My Code Go?

```dot
digraph code_placement {
    rankdir=TB;
    node [shape=diamond];
    "Used by both\nfrontend & backend?" -> "src/shared/" [label="yes"];
    "Used by both\nfrontend & backend?" -> "Runs in\nNode.js only?" [label="no"];
    "Runs in\nNode.js only?" -> "IPC handler?" [label="yes"];
    "Runs in\nNode.js only?" -> "React component\nor hook?" [label="no"];
    "IPC handler?" -> "src/backend/handlers/" [label="yes"];
    "IPC handler?" -> "src/backend/services/" [label="no"];
    "React component\nor hook?" -> "src/frontend/components/" [label="component"];
    "React component\nor hook?" -> "src/frontend/hooks/" [label="hook"];
    "React component\nor hook?" -> "src/frontend/services/" [label="service/singleton"];

    node [shape=box];
    "src/shared/" [style=filled, fillcolor="#e8f5e9"];
    "src/backend/handlers/" [style=filled, fillcolor="#e3f2fd"];
    "src/backend/services/" [style=filled, fillcolor="#e3f2fd"];
    "src/frontend/components/" [style=filled, fillcolor="#fff3e0"];
    "src/frontend/hooks/" [style=filled, fillcolor="#fff3e0"];
    "src/frontend/services/" [style=filled, fillcolor="#fff3e0"];
}
```

**Shared code constraint**: `src/shared/` must NOT import browser APIs, Node.js APIs, or React.

## How Do I Access / Mutate Data?

```dot
digraph data_access {
    rankdir=TB;
    node [shape=diamond];
    "Inside React\ncomponent?" -> "Use query hooks" [label="yes"];
    "Inside React\ncomponent?" -> "Use queryService\nsingleton" [label="no\n(command, service)"];

    "Need to mutate\nfile tree?" -> "queryService.mutations.*\n(optimistic updates)" [label="always"];
    "Need to read\nfile content?" -> "readFileContent()\nhelper (cache-first)" [label="preferred"];

    node [shape=box];
    "Use query hooks" [style=filled, fillcolor="#e8f5e9"];
    "Use queryService\nsingleton" [style=filled, fillcolor="#e8f5e9"];
    "queryService.mutations.*\n(optimistic updates)" [style=filled, fillcolor="#fff3e0"];
    "readFileContent()\nhelper (cache-first)" [style=filled, fillcolor="#fff3e0"];
}
```

**Never call `window.api.*` directly for tree mutations** — always go through `queryService.mutations.*` which handles optimistic updates and rollback.

## Key Directories

| Directory | What to find |
|---|---|
| `src/types.ts` | All shared types (FileNode, FileMetadata, ChatSession, Agent, etc.) |
| `src/shared/` | Code shared by both processes (graph, agentFlow, toolDescriptors, search) |
| `src/backend/handlers/` | IPC handlers: file, project, agent, agentFlow |
| `src/backend/services/` | IndexedFsService, UserDataService, ReferenceIndexService, agent/ |
| `src/backend/services/agent/` | AI system: AdkRunner, ToolFactory, providers, ChatDatabaseService |
| `src/frontend/components/` | All React components (chat/, graph/, fileViewer/, collectionViewer/) |
| `src/frontend/services/commandInfra/` | VS Code-style commands, keybindings, context keys |
| `src/frontend/services/query/` | TanStack Query: queryService, mutations, treeTransformations |
| `src/frontend/hooks/` | Custom hooks: useFileTree, useChat, useAgentFlowRun, etc. |
| `src/frontend/styles/plate/` | Plate.js editor plugins and kits |
| `src/locales/{en,zh-CN}/` | i18n translations (8 namespaces) |
| `e2e/` | Playwright E2E tests |
| `electron/preload.ts` | IPC bridge — `window.api.*` definitions |
| `src/global.d.ts` | TypeScript declarations for `Window.api` |

## Common Task Recipes

Read `references/recipes.md` for **detailed step-by-step procedures** including:
- Adding a new IPC channel (5-step checklist)
- Adding a new command (5-step checklist)
- Adding a new file subType (8-step checklist)
- Adding a new agent tool (6-step checklist)
- Adding a new sidebar panel or editor handler
- Adding i18n strings

## Architecture at a Glance

Read `references/architecture.md` for the full breakdown. Key mental model:

```
Frontend (React) --window.api.*--> Preload --IPC--> Backend handlers --> Services --> Disk
                 <--transport push events--  <--IPC--  ProjectManager <-- IndexedFsService events
```

- **State**: TanStack Query is single source of truth. `staleTime: Infinity` everywhere.
- **File tree**: Event-driven push from backend, never polled. `useFileTree` sets up listener.
- **Mutations**: `createTreeMutation` factory: optimistic update -> IPC call -> rollback on error.
- **Commands**: VS Code-style `CommandService` + `ContextService` + `KeybindingService`.
- **Editor routing**: `ContentViewer` dispatches by `subType` first, then `category`.
- **AI**: Google ADK with multi-provider LLM registry. See `references/agent-system.md`.

## i18n Checklist

- 8 namespaces: `common`, `fileExplorer`, `chat`, `project`, `editor`, `agentBuilder`, `graph`, `agentDefinition`
- In components: `const { t } = useTranslation("namespace")`
- Outside React: `import i18n from "@/src/i18n/i18n"; i18n.t("ns:key")`
- **Always add keys to BOTH `en` and `zh-CN`**

## Build, Lint & Test

```bash
npm run dev:electron     # Dev mode (Vite + Electron)
npm run build            # Production build
npm run check            # Lint + format + TypeScript check
npm run test:e2e         # Playwright E2E tests (headless)
npm run test:e2e:headed  # E2E with visible browser
```

- Path alias: `@/*` maps to repo root
- Formatting: Prettier (double quotes, 4-space indent, semicolons)
- **Always run E2E tests you wrote before committing**

## Pitfalls

1. **Any node can have children** — not just folders
2. **parentId/siblingIndex are derived** — never persist
3. **Tree mutations must be pure** — `treeTransformations.ts` returns new arrays, never mutate
4. **Always add i18n to BOTH locales** — `en` and `zh-CN`
5. **Toolbar renders via portal** — `#file-toolbar-portal` in TopBar, not inside editor DOM
6. **Input elements suppress keybindings** — `useKeybindings` checks `event.target` tag
7. **subType controls routing** — checked before category in ContentViewer
8. **Binary files use base64 over IPC** — text uses UTF-8
9. **STRUCTURAL_FILE_CHANGES control broadcasts** — content "update" does NOT re-push tree
10. **Add Window.api types** — update `src/global.d.ts` when adding IPC channels
11. **Shared code is process-agnostic** — `src/shared/` cannot import browser or Node APIs
12. **Tool descriptors are shared** — `toolDescriptors.ts` serves both UI commands and LLM agents
13. **Every user-facing data element needs a tooltip**

## References

| Reference | Use when |
|---|---|
| `references/architecture.md` | Understanding layers, IPC channels, backend services, frontend providers |
| `references/command-system.md` | Working with commands, keybindings, context keys, when clauses |
| `references/workspace-structure.md` | On-disk file layout, metadata schema, storage guidelines |
| `references/agent-system.md` | AI features: providers, streaming, tools, planning mode, chat persistence |
| `references/recipes.md` | Step-by-step procedures for common development tasks |

Related Skills

---

3891
from openclaw/skills

name: article-factory-wechat

Content & Documentation

humanizer

3891
from openclaw/skills

Remove signs of AI-generated writing from text. Use when editing or reviewing text to make it sound more natural and human-written. Based on Wikipedia's comprehensive "Signs of AI writing" guide. Detects and fixes patterns including: inflated symbolism, promotional language, superficial -ing analyses, vague attributions, em dash overuse, rule of three, AI vocabulary words, negative parallelisms, and excessive conjunctive phrases.

Content & Documentation

find-skills

3891
from openclaw/skills

Helps users discover and install agent skills when they ask questions like "how do I do X", "find a skill for X", "is there a skill that can...", or express interest in extending capabilities. This skill should be used when the user is looking for functionality that might exist as an installable skill.

General Utilities

tavily-search

3891
from openclaw/skills

Use Tavily API for real-time web search and content extraction. Use when: user needs real-time web search results, research, or current information from the web. Requires Tavily API key.

Data & Research

baidu-search

3891
from openclaw/skills

Search the web using Baidu AI Search Engine (BDSE). Use for live information, documentation, or research topics.

Data & Research

agent-autonomy-kit

3891
from openclaw/skills

Stop waiting for prompts. Keep working.

Workflow & Productivity

Meeting Prep

3891
from openclaw/skills

Never walk into a meeting unprepared again. Your agent researches all attendees before calendar events—pulling LinkedIn profiles, recent company news, mutual connections, and conversation starters. Generates a briefing doc with talking points, icebreakers, and context so you show up informed and confident. Triggered automatically before meetings or on-demand. Configure research depth, advance timing, and output format. Walking into meetings blind is amateur hour—missed connections, generic small talk, zero leverage. Use when setting up meeting intelligence, researching specific attendees, generating pre-meeting briefs, or automating your prep workflow.

Workflow & Productivity

self-improvement

3891
from openclaw/skills

Captures learnings, errors, and corrections to enable continuous improvement. Use when: (1) A command or operation fails unexpectedly, (2) User corrects Claude ('No, that's wrong...', 'Actually...'), (3) User requests a capability that doesn't exist, (4) An external API or tool fails, (5) Claude realizes its knowledge is outdated or incorrect, (6) A better approach is discovered for a recurring task. Also review learnings before major tasks.

Agent Intelligence & Learning

botlearn-healthcheck

3891
from openclaw/skills

botlearn-healthcheck — BotLearn autonomous health inspector for OpenClaw instances across 5 domains (hardware, config, security, skills, autonomy); triggers on system check, health report, diagnostics, or scheduled heartbeat inspection.

DevOps & Infrastructure

linkedin-cli

3891
from openclaw/skills

A bird-like LinkedIn CLI for searching profiles, checking messages, and summarizing your feed using session cookies.

Content & Documentation

notebooklm

3891
from openclaw/skills

Google NotebookLM 非官方 Python API 的 OpenClaw Skill。支持内容生成(播客、视频、幻灯片、测验、思维导图等)、文档管理和研究自动化。当用户需要使用 NotebookLM 生成音频概述、视频、学习材料或管理知识库时触发。

Data & Research

小红书长图文发布 Skill

3891
from openclaw/skills

## 概述

Content & Documentation