### Setup Ollama

Source: https://github.com/matevip/mateclaw/blob/dev/mateclaw-server/src/main/resources/docs/en/models.md

Example setup command for Ollama. This is a placeholder and requires actual command to be provided.

```bash
Setup:

```bash

```

--------------------------------

### Full Production Build Sequence

Source: https://github.com/matevip/mateclaw/blob/dev/mateclaw-server/src/main/resources/docs/en/desktop.md

A step-by-step guide to building the production installer. This involves building the frontend, backend JAR, copying resources, downloading the JRE, and finally building the installer.

```bash
# 1. Build frontend static assets
cd mateclaw-ui
pnpm install && pnpm build

# 2. Build backend JAR (includes frontend assets in static/)
cd ../mateclaw-server
mvn clean package -DskipTests

# 3. Copy JAR to desktop resources
JAR_FILE=$(ls -1 target/mateclaw-server-*.jar | grep -v sources | head -n 1)
cp "$JAR_FILE" ../mateclaw-desktop/resources/app.jar

# 4. Download platform-specific JRE
cd ../mateclaw-desktop
bash scripts/download-jre.sh

# 5. Build the installer
pnpm build && npx electron-builder
```

--------------------------------

### Basic Presentation Setup

Source: https://github.com/matevip/mateclaw/blob/dev/mateclaw-server/src/main/resources/skills/pptx/pptxgenjs.md

Initializes a presentation, sets layout and metadata, adds a slide with text, and saves the file. Ensure 'pptxgenjs' is installed.

```javascript
const pptxgen = require("pptxgenjs");

let pres = new pptxgen();
pres.layout = 'LAYOUT_16x9';  // or 'LAYOUT_16x10', 'LAYOUT_4x3', 'LAYOUT_WIDE'
pres.author = 'Your Name';
pres.title = 'Presentation Title';

let slide = pres.addSlide();
slide.addText("Hello World!", { x: 0.5, y: 0.5, fontSize: 36, color: "363636" });

pres.writeFile({ fileName: "Presentation.pptx" });
```

--------------------------------

### Get Setup Status

Source: https://github.com/matevip/mateclaw/blob/dev/mateclaw-server/src/main/resources/docs/en/api.md

Retrieves the general setup status.

```APIDOC
## GET /api/v1/setup/status

### Description
Gets the general setup status.

### Method
GET

### Endpoint
/api/v1/setup/status
```

--------------------------------

### Init Setup

Source: https://github.com/matevip/mateclaw/blob/dev/mateclaw-server/src/main/resources/docs/en/api.md

Initializes the first-run setup process.

```APIDOC
## POST /api/v1/setup/init

### Description
Initializes the first-run setup process.

### Method
POST

### Endpoint
/api/v1/setup/init
```

--------------------------------

### Start the MateClaw Frontend

Source: https://github.com/matevip/mateclaw/blob/dev/mateclaw-server/src/main/resources/docs/en/contributing.md

Install frontend dependencies using pnpm and start the development server. The frontend runs on port 5173 and proxies API requests to the backend.

```bash
cd mateclaw-ui
pnpm install
pnpm dev
```

--------------------------------

### Agent Memory Search Examples

Source: https://github.com/matevip/mateclaw/blob/dev/mateclaw-server/src/main/resources/docs/zh/memory.md

Examples demonstrating how to search an agent's memory for specific files or content. The 'list' example shows how to get a list of files, while 'read' and 'edit' show how to access and modify file content.

```json
// Input
{"agentId": 1, "filenamePrefix": "memory/"}
// Output
{"agentId": 1, "count": 3, "files": [
  {"filename": "memory/2026-04-09.md", "enabled": false, "fileSize": 512},
  ...
]}
```

```json
// Input
{"agentId": 1, "filename": "MEMORY.md"}
// Output
{"agentId": 1, "filename": "MEMORY.md", "enabled": true, "content": "..."}
```

```json
// Input
{"agentId": 1, "filename": "MEMORY.md", "oldText": "旧内容", "newText": "新内容"}
// Output
{"agentId": 1, "filename": "MEMORY.md", "replacements": 1}
```

--------------------------------

### Development Mode Commands

Source: https://github.com/matevip/mateclaw/blob/dev/mateclaw-server/src/main/resources/docs/en/desktop.md

Navigate to the desktop directory, install dependencies, and start the development server. This launches the frontend dev server, Electron main process, and spawns the Spring Boot backend.

```bash
cd mateclaw-desktop
pnpm install
pnpm dev
```

--------------------------------

### Install ascii-image-converter via go install

Source: https://github.com/matevip/mateclaw/blob/dev/mateclaw-server/src/main/resources/skills/ascii-art/SKILL.md

Installs the 'ascii-image-converter' tool using 'go install'. Ensure Go is installed and configured.

```bash
go install github.com/TheZoraiz/ascii-image-converter@latest
```

--------------------------------

### Clone the MateClaw Repository

Source: https://github.com/matevip/mateclaw/blob/dev/mateclaw-server/src/main/resources/docs/en/contributing.md

Start by forking the MateClaw repository and cloning your fork to your local machine. Navigate into the cloned directory to begin setup.

```bash
git clone https://github.com/YOUR_USERNAME/mateclaw.git
cd mateclaw
```

--------------------------------

### Install Boxes

Source: https://github.com/matevip/mateclaw/blob/dev/mateclaw-server/src/main/resources/skills/ascii-art/SKILL.md

Install the boxes utility for drawing decorative ASCII art borders around text. Installation commands vary by OS.

```bash
sudo apt install boxes -y    # Debian/Ubuntu
# brew install boxes         # macOS
```

--------------------------------

### Install Outlines Base and Backends

Source: https://github.com/matevip/mateclaw/blob/dev/mateclaw-server/src/main/resources/skills/outlines/SKILL.md

Install the base Outlines library and optional backends for specific model integrations like Transformers, llama.cpp, or vLLM.

```bash
pip install outlines

pip install outlines transformers  # Hugging Face models
pip install outlines llama-cpp-python  # llama.cpp
pip install outlines vllm  # vLLM for high-throughput
```

--------------------------------

### Docker: Deploy vLLM with Outlines

Source: https://github.com/matevip/mateclaw/blob/dev/mateclaw-server/src/main/resources/skills/outlines/references/backends.md

Create a Docker image for serving models with vLLM and Outlines. This example uses the official vLLM-OpenAI image and installs Outlines.

```dockerfile
FROM vllm/vllm-openai:latest

# Install outlines
RUN pip install outlines

# Copy your code
COPY app.py /app/

# Run
CMD ["python", "/app/app.py"]
```

--------------------------------

### Setup Webhook Platform Wizard

Source: https://github.com/matevip/mateclaw/blob/dev/mateclaw-server/src/main/resources/skills/webhook-subscriptions/SKILL.md

Initiates an interactive setup wizard to enable the webhook platform, configure the port, and set a global HMAC secret.

```bash
the agent gateway setup
```

--------------------------------

### Install Cowsay

Source: https://github.com/matevip/mateclaw/blob/dev/mateclaw-server/src/main/resources/skills/ascii-art/SKILL.md

Install the cowsay utility for creating message art with ASCII characters and speech bubbles. Installation commands vary by OS.

```bash
sudo apt install cowsay -y    # Debian/Ubuntu
# brew install cowsay         # macOS
```

--------------------------------

### Install ascii-image-converter via snap

Source: https://github.com/matevip/mateclaw/blob/dev/mateclaw-server/src/main/resources/skills/ascii-art/SKILL.md

Installs the 'ascii-image-converter' tool using the snap package manager.

```bash
sudo snap install ascii-image-converter
```

--------------------------------

### preload(), setup(), draw()

Source: https://github.com/matevip/mateclaw/blob/dev/mateclaw-server/src/main/resources/skills/p5js/references/core-api.md

The fundamental p5.js functions that control the program's lifecycle: `preload` for loading assets, `setup` for initialization, and `draw` for rendering each frame.

```APIDOC
## preload()

### Description
Loads assets (images, fonts, data) before the sketch starts. Execution is blocked until all assets are loaded.

### Example
```javascript
function preload() {
  font = loadFont('font.otf');
  img = loadImage('texture.png');
  data = loadJSON('data.json');
}
```

## setup()

### Description
Runs once at the beginning of the sketch. Used for initial setup like creating the canvas and setting initial states.

### Example
```javascript
function setup() {
  createCanvas(1920, 1080);
  colorMode(HSB, 360, 100, 100, 100);
  randomSeed(CONFIG.seed);
  noiseSeed(CONFIG.seed);
}
```

## draw()

### Description
Runs repeatedly every frame (default 60fps) to draw the sketch. Can be controlled with `frameRate()`, `noLoop()`, `loop()`, and `redraw()`.

### Example
```javascript
function draw() {
  // Runs every frame (default 60fps).
  // Set frameRate(30) to change.
  // Call noLoop() for static sketches (render once).
}
```
```

--------------------------------

### Blogwatcher CLI Example Output: Scan

Source: https://github.com/matevip/mateclaw/blob/dev/mateclaw-server/src/main/resources/skills/blogwatcher/SKILL.md

Example output after scanning blogs. Indicates the number of blogs scanned, articles found, and new articles discovered.

```text
$ blogwatcher-cli scan
Scanning 1 blog(s)...

  xkcd
    Source: RSS | Found: 4 | New: 4

Found 4 new article(s) total!
```

--------------------------------

### Install pyfiglet

Source: https://github.com/matevip/mateclaw/blob/dev/mateclaw-server/src/main/resources/skills/ascii-art/SKILL.md

Install the pyfiglet library for local ASCII art banner generation. Use --break-system-packages for isolated environments.

```bash
pip install pyfiglet --break-system-packages -q
```

--------------------------------

### p5.js 1.x vs 2.x Asynchronous Setup

Source: https://github.com/matevip/mateclaw/blob/dev/mateclaw-server/src/main/resources/skills/p5js/references/core-api.md

Demonstrates the change from `preload()` in p5.js 1.x to `async setup()` in p5.js 2.x for handling asynchronous operations like image loading.

```javascript
// p5.js 1.x
let img;
function preload() { img = loadImage('cat.jpg'); }
function setup() { createCanvas(800, 800); }

// p5.js 2.x
let img;
async function setup() {
  createCanvas(800, 800);
  img = await loadImage('cat.jpg');
}
```

--------------------------------

### Research Code README Template

Source: https://github.com/matevip/mateclaw/blob/dev/mateclaw-server/src/main/resources/skills/research-paper-writing/SKILL.md

A template for the README.md file in a research code repository, guiding users on setup, reproduction, and citation.

```markdown
# [Paper Title]

Official implementation of "[Paper Title]" (Venue Year).

## Setup
[Exact commands to set up environment]

## Reproduction
To reproduce Table 1: `bash scripts/reproduce_table1.sh`
To reproduce Figure 2: `python scripts/make_figure2.py`

## Citation
[BibTeX entry]
```

--------------------------------

### Basic Document Setup

Source: https://github.com/matevip/mateclaw/blob/dev/mateclaw-server/src/main/resources/skills/docx/SKILL.md

Initializes a new Document object and exports it to a DOCX file. Requires importing necessary components from 'docx'.

```javascript
const { Document, Packer, Paragraph, TextRun, Table, TableRow, TableCell, ImageRun,
        Header, Footer, AlignmentType, PageOrientation, LevelFormat, ExternalHyperlink,
        TableOfContents, HeadingLevel, BorderStyle, WidthType, ShadingType,
        VerticalAlign, PageNumber, PageBreak } = require('docx');

const doc = new Document({ sections: [{ children: [/* content */] }] });
Packer.toBuffer(doc).then(buffer => fs.writeFileSync("doc.docx", buffer));
```

--------------------------------

### Raw v2 API Endpoint Access

Source: https://github.com/matevip/mateclaw/blob/dev/mateclaw-server/src/main/resources/skills/x_intel/SKILL.md

Examples of directly accessing X API v2 GET endpoints for fetching user information, tweets with expansions, and performing recent searches. Includes examples for public user fields and tweet details.

```bash
xurl /2/users/by/username/elonmusk?user.fields=public_metrics,description,verified
```

```bash
xurl /2/tweets/1234567890?tweet.fields=public_metrics,created_at&expansions=author_id
```

```bash
xurl /2/tweets/search/recent?query=langchain&tweet.fields=created_at,public_metrics&max_results=25
```

```bash
xurl https://api.x.com/2/users/me
```

--------------------------------

### Example: Filesystem MCP (stdio)

Source: https://github.com/matevip/mateclaw/blob/dev/mateclaw-server/src/main/resources/docs/en/mcp.md

Demonstrates creating a Filesystem MCP server using stdio transport. This server provides tools for file system operations within a specified directory.

```bash
curl -X POST http://localhost:18088/api/v1/mcp/servers \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer <token>" \
  -d '{
    "name": "filesystem",
    "description": "Filesystem access (restricted to specified directory)",
    "transport": "stdio",
    "command": "npx",
    "argsJson": "[\"-y\", \"@anthropic/mcp-filesystem\", \"/home/user/workspace\"]",
    "enabled": true
  }'
```

--------------------------------

### Initial Mateclaw Docker Deployment Steps

Source: https://github.com/matevip/mateclaw/blob/dev/mateclaw-server/src/main/resources/docs/en/docker-deploy.md

Commands to clone the Mateclaw repository, navigate into the directory, and copy the example environment file. This is the first step in setting up the Docker deployment.

```sh
git clone https://github.com/matevip/mateclaw.git
cd mateclaw

# 1. Fill in required values
cp .env.example .env
vi .env   # see table below
```

--------------------------------

### Common Song Structures

Source: https://github.com/matevip/mateclaw/blob/dev/mateclaw-server/src/main/resources/skills/songwriting-and-ai-music/SKILL.md

Examples of common song structures used in various music genres. These serve as starting points and can be modified or combined.

```text
ABABCB  Verse/Chorus/Verse/Chorus/Bridge/Chorus    (most pop/rock)
AABA    Verse/Verse/Bridge/Verse (refrain-based)    (jazz standards, ballads)
ABAB    Verse/Chorus alternating                    (simple, direct)
AAA     Verse/Verse/Verse (strophic, no chorus)     (folk, storytelling)
```

--------------------------------

### Explore Repository Command

Source: https://github.com/matevip/mateclaw/blob/dev/mateclaw-server/src/main/resources/skills/research-paper-writing/SKILL.md

Use this bash command to begin exploring the project repository as part of the initial setup phase.

```bash
git ls-tree -r HEAD
```

--------------------------------

### Iterative Design Steps

Source: https://github.com/matevip/mateclaw/blob/dev/mateclaw-server/src/main/resources/skills/popular-web-designs/templates/sanity.md

A step-by-step guide for iteratively building a design, starting with a dark theme and progressively adding structure, typography, color, and details.

```text
1. Start dark: Begin with #0b0b0b background, #ffffff primary text, #b9b9b9 secondary text
2. Add structure: Use #212121 surfaces and #353535 borders for containment -- no shadows
3. Apply typography: Inter (or Space Grotesk) with tight letter-spacing on headings, 1.50 line-height on body
4. Color punctuation: Add #f36458 for CTAs and #0052ef for all hover/interactive states
5. Refine spacing: 8px base unit, 24-32px within sections, 64-120px between sections
6. Technical details: Add IBM Plex Mono uppercase labels for tags and metadata
7. Polish: Ensure all interactive elements hover to #0052ef, all buttons are pills or subtle 5px radius, borders are hairline (1px)
```

--------------------------------

### Basic vLLM Setup

Source: https://github.com/matevip/mateclaw/blob/dev/mateclaw-server/src/main/resources/skills/outlines/references/backends.md

Loads a model using the vLLM backend and prepares it for JSON generation. Ensure the model name is compatible with vLLM.

```python
import outlines

# Load model with vLLM
model = outlines.models.vllm("meta-llama/Llama-3.1-8B-Instruct")

# Use with generator
generator = outlines.generate.json(model, YourModel)
```

--------------------------------

### Development Server Commands

Source: https://github.com/matevip/mateclaw/blob/dev/mateclaw-server/src/main/resources/docs/en/console.md

Provides essential commands for setting up and running the development environment for the mateclaw-ui project. Includes installation, starting the dev server, building, and linting.

```bash
cd mateclaw-ui
pnpm install
pnpm dev          # Port 5173, proxies /api to :18088
pnpm build        # vue-tsc + vite build into ../mateclaw-server/.../static
pnpm lint         # ESLint
```

--------------------------------

### Setup p5.js Skill Dependencies

Source: https://github.com/matevip/mateclaw/blob/dev/mateclaw-server/src/main/resources/skills/p5js/README.md

Run this script to verify and set up necessary dependencies for the p5.js skill, particularly for headless export.

```bash
bash skills/creative/p5js/scripts/setup.sh
```

--------------------------------

### Vue Component Structure Example

Source: https://github.com/matevip/mateclaw/blob/dev/mateclaw-server/src/main/resources/docs/en/contributing.md

A standard Vue 3 component using `<script setup>` and Composition API, integrating with Pinia for state management and Element Plus for UI elements.

```vue
<script setup lang="ts">
import { ref, onMounted } from 'vue'
import { useAgentStore } from '@/stores/useAgentStore'

const agentStore = useAgentStore()
const loading = ref(false)

onMounted(async () => {
  loading.value = true
  await agentStore.fetchAgents()
  loading.value = false
})
</script>

<template>
  <div class="p-4">
    <el-table :data="agentStore.agents" v-loading="loading">
      <!-- columns -->
    </el-table>
  </div>
</template>
```

--------------------------------

### Create and Use ReAct Agent

Source: https://github.com/matevip/mateclaw/blob/dev/mateclaw-server/src/main/resources/skills/dspy/references/examples.md

Demonstrates how to initialize a ReAct agent with specific tools and then use it to answer a question that requires tool usage and calculation.

```python
from dspy.agents import ReAct
from dspy.tools import search_wikipedia, calculate, search_web

# Define a dummy ResearchAgent for the example
class ResearchAgent:
    def __call__(self, query):
        if "population of France" in query:
            return "67,000,000"
        return ""

# Create ReAct agent
agent = ReAct(ResearchAgent, tools=[search_wikipedia, calculate, search_web])

# Agent decides which tools to use
question = "What is the population of France divided by 10?"
result = agent(question=question)
# Agent:
# 1. Thinks: "Need population of France"
# 2. Acts: search_wikipedia("France population")
# 3. Thinks: "Got 67 million, need to divide"
# 4. Acts: calculate("67000000 / 10")
# 5. Returns: "6,700,000"
```

--------------------------------

### Code Typing Start Time

Source: https://github.com/matevip/mateclaw/blob/dev/mateclaw-server/src/main/resources/skills/pretext/templates/donut-orbit.html

Calculates the start time for code typing animation, relative to the orb build start time.

```javascript
function codeTypingStartTime() {
  return Math.max(0, orbBuildStartTime() + SCENE.orbBuildDuration * 0.62 - 0.3);
}
```

--------------------------------

### Microphone Input and Audio Analysis in p5.js

Source: https://github.com/matevip/mateclaw/blob/dev/mateclaw-server/src/main/resources/skills/p5js/references/interaction.md

Sets up microphone input and initializes p5.sound objects for analyzing audio. It demonstrates how to start the audio context with user gesture, create an AudioIn object, and configure FFT and Amplitude analyzers to process sound data.

```javascript
let mic, fft, amplitude;

function setup() {
  createCanvas(800, 800);
  userStartAudio();  // required — user gesture to enable audio

  mic = new p5.AudioIn();
  mic.start();

  fft = new p5.FFT(0.8, 256);  // smoothing, bins
  fft.setInput(mic);

  amplitude = new p5.Amplitude();
  amplitude.setInput(mic);
}

function draw() {
  let level = amplitude.getLevel();    // 0.0 to 1.0 (overall volume)
  let spectrum = fft.analyze();         // array of 256 frequency values (0-255)
  let waveform = fft.waveform();        // array of 256 time-domain samples (-1 to 1)

  // Get energy in frequency bands
  let bass = fft.getEnergy('bass');          // 20-140 Hz
  let lowMid = fft.getEnergy('lowMid');      // 140-400 Hz
  let mid = fft.getEnergy('mid');            // 400-2600 Hz
  let highMid = fft.getEnergy('highMid');    // 2600-5200 Hz
  let treble = fft.getEnergy('treble');      // 5200-14000 Hz
  // Each returns 0-255
}
```

--------------------------------

### Basic OpenAI Model Setup

Source: https://github.com/matevip/mateclaw/blob/dev/mateclaw-server/src/main/resources/skills/outlines/references/backends.md

Sets up a basic connection to an OpenAI model using the `outlines.models.openai` function. Requires an API key and specifies the model name.

```python
import outlines

# Basic OpenAI support
model = outlines.models.openai("gpt-4o-mini", api_key="your-api-key")

# Use with generator
generator = outlines.generate.json(model, YourModel)
result = generator("Your prompt")
```

--------------------------------

### Build MateClaw Frontend

Source: https://github.com/matevip/mateclaw/blob/dev/mateclaw-server/src/main/resources/docs/en/faq.md

Navigate to the `mateclaw-ui` directory and run `pnpm build` to build the frontend. Verify the output in the server's static resources.

```bash
cd mateclaw-ui
pnpm build
ls ../mateclaw-server/src/main/resources/static/
```

--------------------------------

### Install DSPy

Source: https://github.com/matevip/mateclaw/blob/dev/mateclaw-server/src/main/resources/skills/dspy/SKILL.md

Install the DSPy library using pip. You can install the stable release, the latest development version, or with specific LM provider support.

```bash
pip install dspy
```

```bash
pip install git+https://github.com/stanfordnlp/dspy.git
```

```bash
pip install dspy[openai]
```

```bash
pip install dspy[anthropic]
```

```bash
pip install dspy[all]
```

--------------------------------

### ReAct Agent with Tools

Source: https://github.com/matevip/mateclaw/blob/dev/mateclaw-server/src/main/resources/skills/dspy/references/examples.md

Demonstrates the setup for a ReAct agent using DSPy. It defines available tools (search_wikipedia, calculate, search_web) and a signature for the agent to interact with these tools to answer questions.

```python
from dspy.predict import ReAct

# Define tools
def search_wikipedia(query: str) -> str:
    """Search Wikipedia for information."""
    import wikipedia
    try:
        return wikipedia.summary(query, sentences=3)
    except:
        return "No results found"

def calculate(expression: str) -> str:
    """Evaluate mathematical expression safely."""
    try:
        # Use safe eval
        result = eval(expression, {"__builtins__": {}}, {})
        return str(result)
    except:
        return "Invalid expression"

def search_web(query: str) -> str:
    """Search the web."""
    # Your web search implementation
    return results

# Create agent signature
class ResearchAgent(dspy.Signature):
    """Answer questions using available tools."""
    question = dspy.InputField()
    answer = dspy.OutputField()
```

--------------------------------

### Install xurl CLI using curl

Source: https://github.com/matevip/mateclaw/blob/dev/mateclaw-server/src/main/resources/skills/x_intel/SKILL.md

Installs the xurl CLI on Linux and macOS systems using a curl script. The installation places the executable in ~/.local/bin and does not require sudo privileges.

```bash
# Shell script (Linux + macOS, installs to ~/.local/bin, no sudo)
curl -fsSL https://raw.githubusercontent.com/xdevplatform/xurl/main/install.sh | bash
```

--------------------------------

### KNNFewShot Optimizer for Example Selection

Source: https://github.com/matevip/mateclaw/blob/dev/mateclaw-server/src/main/resources/skills/dspy/references/optimizers.md

Employ KNNFewShot for a simple k-nearest neighbors approach to select similar examples for few-shot learning. It's a good baseline when training examples are diverse.

```python
from dspy.teleprompt import KNNFewShot

trainset = [...] # Diverse examples recommended

# No metric needed - just selects similar examples
optimizer = KNNFewShot(k=3)
optimized_qa = optimizer.compile(qa, trainset=trainset)

# For each query, uses 3 most similar examples from trainset
```

--------------------------------

### Install latexdiff

Source: https://github.com/matevip/mateclaw/blob/dev/mateclaw-server/src/main/resources/skills/research-paper-writing/SKILL.md

Provides installation instructions for latexdiff on macOS and Linux systems.

```bash
# Install
# macOS: brew install latexdiff (or comes with TeX Live)
# Linux: sudo apt install latexdiff
```

--------------------------------

### State Management Example

Source: https://github.com/matevip/mateclaw/blob/dev/mateclaw-server/src/main/resources/docs/en/console.md

Demonstrates correct and incorrect ways to interact with the agent and theme stores. Always use store actions for fetching and setting data, and avoid direct mutation of state properties.

```typescript
// Correct — go through store actions
agentStore.fetchAgents()
themeStore.setMode('dark')

// Wrong — never mutate directly
agentStore.agents = []         // Don't
```

--------------------------------

### Install SciencePlots

Source: https://github.com/matevip/mateclaw/blob/dev/mateclaw-server/src/main/resources/skills/research-paper-writing/SKILL.md

Installs the SciencePlots library for enhancing matplotlib plots with publication-quality styles.

```bash
pip install SciencePlots
```

--------------------------------

### List Available OpenAI Models

Source: https://github.com/matevip/mateclaw/blob/dev/mateclaw-server/src/main/resources/skills/outlines/references/backends.md

Shows examples of how to instantiate Outlines with different available OpenAI models, including GPT-4o, GPT-4o Mini, GPT-4 Turbo, and GPT-3.5 Turbo.

```python
# GPT-4o (latest)
model = outlines.models.openai("gpt-4o")

# GPT-4o Mini (cost-effective)
model = outlines.models.openai("gpt-4o-mini")

# GPT-4 Turbo
model = outlines.models.openai("gpt-4-turbo")

# GPT-3.5 Turbo
model = outlines.models.openai("gpt-3.5-turbo")
```

--------------------------------

### Install jp2a

Source: https://github.com/matevip/mateclaw/blob/dev/mateclaw-server/src/main/resources/skills/ascii-art/SKILL.md

Installs the 'jp2a' tool, a lightweight JPEG to ASCII converter, on Debian/Ubuntu systems.

```bash
sudo apt install jp2a -y
```

--------------------------------

### Example Workflow for Task Management

Source: https://github.com/matevip/mateclaw/blob/dev/mateclaw-server/src/main/resources/skills/subagent-driven-development/SKILL.md

An illustrative workflow demonstrating task execution, review, and completion using subagents.

```bash
[Read plan: docs/plans/auth-feature.md]
[Create todo list with 5 tasks]

--- Task 1: Create User model ---
[Dispatch implementer subagent]
  Implementer: "Should email be unique?"
  You: "Yes, email must be unique"
  Implementer: Implemented, 3/3 tests passing, committed.

[Dispatch spec reviewer]
  Spec reviewer: ✅ PASS — all requirements met

[Dispatch quality reviewer]
  Quality reviewer: ✅ APPROVED — clean code, good tests

[Mark Task 1 complete]

--- Task 2: Password hashing ---
[Dispatch implementer subagent]
  Implementer: No questions, implemented, 5/5 tests passing.

[Dispatch spec reviewer]
  Spec reviewer: ❌ Missing: password strength validation (spec says "min 8 chars")

[Implementer fixes]
  Implementer: Added validation, 7/7 tests passing.

[Dispatch spec reviewer again]
  Spec reviewer: ✅ PASS

[Dispatch quality reviewer]
  Quality reviewer: Important: Magic number 8, extract to constant
  Implementer: Extracted MIN_PASSWORD_LENGTH constant
  Quality reviewer: ✅ APPROVED

[Mark Task 2 complete]

... (continue for all tasks)

[After all tasks: dispatch final integration reviewer]
[Run full test suite: all passing]
[Done!]
```

--------------------------------

### Per-Step Model Configuration Example

Source: https://github.com/matevip/mateclaw/blob/dev/mateclaw-server/src/main/resources/docs/en/wiki.md

Defines different LLM models for specific steps within the ingestion process. This allows for cost optimization and performance tuning by using cheaper models for simpler tasks and stronger models for complex ones.

```text
heavy_ingest.route        → small, cheap model for routing
heavy_ingest.create_page  → strong model for full-page authoring
heavy_ingest.merge_page   → strong model for content merging
light_enrich.enrich       → small, cheap model for wikilink annotation
```

--------------------------------

### Install toilet (Debian/Ubuntu)

Source: https://github.com/matevip/mateclaw/blob/dev/mateclaw-server/src/main/resources/skills/ascii-art/SKILL.md

Installs the 'toilet' command-line tool and its fonts on Debian or Ubuntu systems.

```bash
sudo apt install toilet toilet-fonts -y
```

--------------------------------

### Troubleshooting: Install Missing LaTeX Packages

Source: https://github.com/matevip/mateclaw/blob/dev/mateclaw-server/src/main/resources/skills/research-paper-writing/templates/README.md

Command to install missing LaTeX packages using the TeX Live Package Manager (`tlmgr`). Alternatively, installing the full TeX Live distribution is recommended to avoid such issues.

```bash
# TeX Live package manager
tlmgr install <package-name>

# Or install full distribution to avoid this

```

--------------------------------

### Git Commit Message Example

Source: https://github.com/matevip/mateclaw/blob/dev/mateclaw-server/src/main/resources/skills/research-paper-writing/SKILL.md

Example of a descriptive Git commit message for tracking experiment batches.

```git
Add Monte Carlo constrained results (5 runs, Sonnet 4.6, policy memo task)
Add Haiku baseline comparison: autoreason vs refinement baselines at cheap model tier
```

--------------------------------

### API Reference: Get Trigger Details

Source: https://github.com/matevip/mateclaw/blob/dev/mateclaw-server/src/main/resources/docs/en/triggers.md

GET request to retrieve details for a specific trigger by its ID.

```http
GET | /api/v1/triggers/{id} | Get details
```

--------------------------------

### Run MateClaw Backend and Frontend

Source: https://github.com/matevip/mateclaw/blob/dev/README.md

Starts the MateClaw backend server and the frontend UI. Access the UI at http://localhost:5173.

```bash
# Backend
cd mateclaw-server
mehndi spring-boot:run           # http://localhost:18088

# Frontend
cd mateclaw-ui
pnpm install && pnpm dev      # http://localhost:5173
```

--------------------------------

### Verify xurl installation

Source: https://github.com/matevip/mateclaw/blob/dev/mateclaw-server/src/main/resources/skills/x_intel/SKILL.md

Verifies that the xurl CLI has been installed correctly by checking its help output and authentication status.

```bash
xurl --help
xurl auth status
```

--------------------------------

### Setup Webhook Platform Environment Variables

Source: https://github.com/matevip/mateclaw/blob/dev/mateclaw-server/src/main/resources/skills/webhook-subscriptions/SKILL.md

Configure webhook platform settings using environment variables in the .env file.

```bash
WEBHOOK_ENABLED=true
WEBHOOK_PORT=8644
WEBHOOK_SECRET=generate-a-strong-secret-here
```

--------------------------------

### Install xurl CLI using Go

Source: https://github.com/matevip/mateclaw/blob/dev/mateclaw-server/src/main/resources/skills/x_intel/SKILL.md

Installs the xurl CLI for cross-platform use with the Go programming language.

```bash
# Go (cross-platform)
go install github.com/xdevplatform/xurl@latest
```

--------------------------------

### Install xurl CLI using Homebrew

Source: https://github.com/matevip/mateclaw/blob/dev/mateclaw-server/src/main/resources/skills/x_intel/SKILL.md

Installs the xurl CLI on macOS systems using the Homebrew package manager.

```bash
# Homebrew (macOS)
brew install --cask xdevplatform/tap/xurl
```