### Register and Install Plugins via CLI

Source: https://context7.com/factory-ai/factory-plugins/llms.txt

Commands to add the marketplace repository, install specific plugins, and access the UI.

```bash
# Add the official Factory plugins marketplace
droid plugin marketplace add https://github.com/Factory-AI/factory-plugins

# Install a specific plugin from the marketplace
droid plugin install security-engineer@factory-plugins

# Browse available plugins via the UI
/plugins
```

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

### Start Slidev Dev Server with Custom Entry

Source: https://github.com/factory-ai/factory-plugins/blob/master/plugins/droid-evolved/skills/visual-design/presentations.md

Starts the development server using a specified markdown file as the entry point.

```bash
slidev my-slides.md
```

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

### Install and configure nanobanana

Source: https://github.com/factory-ai/factory-plugins/blob/master/plugins/droid-evolved/skills/visual-design/image-generation.md

Installation instructions and API key configuration for the CLI tool.

```bash
npm install -g @factory/nanobanana
```

```bash
export GEMINI_API_KEY="your-api-key"
```

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

### Install a Plugin from Marketplace

Source: https://github.com/factory-ai/factory-plugins/blob/master/README.md

After adding the marketplace, use this command to install a specific plugin, such as 'security-engineer', from the 'factory-plugins' source.

```bash
droid plugin install security-engineer@factory-plugins
```

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

### Start Slidev Dev Server (Default)

Source: https://github.com/factory-ai/factory-plugins/blob/master/plugins/droid-evolved/skills/visual-design/presentations.md

Starts the development server with default settings, typically using `slides.md` on port 3030.

```bash
slidev
```

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

### Start Slidev Dev Server with Custom Port and Auto-Open

Source: https://github.com/factory-ai/factory-plugins/blob/master/plugins/droid-evolved/skills/visual-design/presentations.md

Starts the development server on a custom port and automatically opens the presentation in the default browser.

```bash
slidev --port 8080 --open
```

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

### Run Slidev Development Server

Source: https://github.com/factory-ai/factory-plugins/blob/master/plugins/droid-evolved/skills/visual-design/presentations.md

Starts the Slidev development server to preview your presentation.

```bash
npx slidev
```

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

### Skill Verification Example

Source: https://github.com/factory-ai/factory-plugins/blob/master/plugins/droid-evolved/skills/skill-creation/SKILL.md

Provides examples of verification steps to ensure a skill has functioned correctly.

```markdown
## Verify it worked

Run `npm test` and make sure nothing broke.
Check that the new file exists at `src/config.ts`.
```

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

### Add Slidev to Existing Project

Source: https://github.com/factory-ai/factory-plugins/blob/master/plugins/droid-evolved/skills/visual-design/presentations.md

Install Slidev CLI and default theme into an existing project.

```bash
npm install @slidev/cli @slidev/theme-default
```

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

### Install agent-browser Globally or via npx

Source: https://github.com/factory-ai/factory-plugins/blob/master/plugins/droid-evolved/skills/browser-navigation/SKILL.md

Install the agent-browser tool globally using npm or execute commands directly using npx.

```bash
npm install -g agent-browser
```

```bash
npx agent-browser open https://example.com
```

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

### Development Commands

Source: https://github.com/factory-ai/factory-plugins/blob/master/plugins/droid-evolved/skills/visual-design/presentations.md

Commands for starting the local development server for Slidev presentations.

```APIDOC
## Development Commands

### Description
Starts the local development server to preview slides.

### Endpoint
`slidev [entry_file]`

### Parameters
#### Path Parameters
- **entry_file** (string) - Optional - The path to the markdown file (default: slides.md)

#### Query Parameters
- **--port** (number) - Optional - Port to run the server on (default: 3030)
- **--open** (boolean) - Optional - Automatically open the browser
- **--remote** (string) - Optional - Enable remote access, optionally with a password
```

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

### Get prompting help

Source: https://github.com/factory-ai/factory-plugins/blob/master/plugins/droid-evolved/skills/visual-design/image-generation.md

Access built-in tips for improving prompts for specific commands.

```bash
nanobanana tips              # general tips
nanobanana tips generate     # tips for generate command
nanobanana tips diagram      # tips for diagram command
```

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

### Scraping Data from a Web Page

Source: https://github.com/factory-ai/factory-plugins/blob/master/plugins/droid-evolved/skills/browser-navigation/SKILL.md

Example of opening a URL, taking a snapshot to get page structure, and then extracting specific text content and attribute values using element references.

```bash
agent-browser open https://example.com/products
agent-browser snapshot -i --json > page_structure.json

# Get specific data
agent-browser get text @e5  # Product title
agent-browser get attr @e6 href  # Product link
```

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

### Form Submission Workflow

Source: https://github.com/factory-ai/factory-plugins/blob/master/plugins/droid-evolved/skills/browser-navigation/SKILL.md

A multi-step example demonstrating how to open a form, identify elements using snapshots, fill in data, click submit, and wait for page load.

```bash
agent-browser open https://example.com/form
snapshots:
agent-browser snapshot -i
# Output shows: textbox "Email" [ref=e1], textbox "Password" [ref=e2], button "Submit" [ref=e3]

agent-browser fill @e1 "user@example.com"
agent-browser fill @e2 "password123"
agent-browser click @e3
agent-browser wait --load networkidle
agent-browser snapshot -i  # Check result
```

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

### Add Marketplace to Factory

Source: https://github.com/factory-ai/factory-plugins/blob/master/README.md

Use this command to add the official Factory plugins marketplace to your Factory installation. This is a prerequisite for installing plugins from this marketplace.

```bash
droid plugin marketplace add https://github.com/Factory-AI/factory-plugins
```

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

### Summary of Created Branches

Source: https://github.com/factory-ai/factory-plugins/blob/master/plugins/autoresearch/skills/autoresearch/SKILL.md

Example output summarizing the creation of clean branches from experiments, including metric improvements and readiness for review. It also notes the preservation of the original experiment branch.

```text
Created 2 clean branches from 20 experiments:

  autoresearch/finalize/reduce-depth
    val_bpb: 1.15 -> 1.08 (-6.1%)
    Ready for review

  autoresearch/finalize/cosine-schedule
    val_bpb: 1.08 -> 1.05 (-2.8%)
    Ready for review

Original experiment branch preserved: autoresearch/<session-branch>
```

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

### Slidev Frontmatter Configuration

Source: https://github.com/factory-ai/factory-plugins/blob/master/plugins/droid-evolved/skills/visual-design/presentations.md

Example of frontmatter for configuring individual slides, including layout and transitions.

```yaml
---
layout: two-cols
transition: slide-left
class: px-20
---
```

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

### Testing a Login Flow

Source: https://github.com/factory-ai/factory-plugins/blob/master/plugins/droid-evolved/skills/browser-navigation/SKILL.md

A comprehensive example for testing a login process, including navigation, filling credentials, clicking submit, waiting for redirection, and verifying successful login by checking for a username element.

```bash
# Navigate to login
agent-browser open https://app.example.com/login

# Take initial snapshot
agent-browser snapshot -i

# Fill credentials
agent-browser fill @e1 "test@example.com"
agent-browser fill @e2 "testpassword"

# Click login
agent-browser click @e3

# Wait for redirect
agent-browser wait --url "**/dashboard"

# Verify logged in
agent-browser get text @e10  # Should show username
```

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

### Enable Remote Access with Password

Source: https://github.com/factory-ai/factory-plugins/blob/master/plugins/droid-evolved/skills/visual-design/presentations.md

Starts the development server with remote access enabled and password protection.

```bash
slidev --remote mypassword
```

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

### Generate and edit images with nanobanana CLI

Source: https://github.com/factory-ai/factory-plugins/blob/master/plugins/droid-evolved/skills/visual-design/SKILL.md

Install the CLI globally and set the API key before running generation or editing commands.

```bash
npm install -g @factory/nanobanana
export GEMINI_API_KEY="your-key"

nanobanana generate "company logo" --count=4 --styles=modern,minimal
nanobanana edit photo.png "remove background"
nanobanana icon "settings gear" --style=flat
nanobanana diagram "auth flow" --type=flowchart
```

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

### Human Writing Transformation Example

Source: https://context7.com/factory-ai/factory-plugins/llms.txt

Comparison of AI-sounding text versus humanized text.

```markdown
# Before (AI-sounding):
The new software update serves as a testament to the company's commitment
to innovation. Moreover, it provides a seamless, intuitive, and powerful
user experience—ensuring that users can accomplish their goals efficiently.

# After (humanized):
The software update adds batch processing, keyboard shortcuts, and offline mode.
Early feedback from beta testers has been positive, with most reporting faster
task completion.
```

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

### Example of Grouping Files in Review

Source: https://github.com/factory-ai/factory-plugins/blob/master/plugins/core/skills/review/SKILL.md

Illustrates how to group modified files into logical clusters for parallel review, based on related functionality, file relationships, risk profile, or dependencies.

```text
- Group 1 (Auth): src/auth/login.ts, src/auth/session.ts, tests/auth.test.ts
- Group 2 (API handlers): src/api/users.ts, src/api/orders.ts
- Group 3 (Database): src/db/migrations/001.ts, src/db/schema.ts
```

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

### Record Browser Sessions as Video

Source: https://github.com/factory-ai/factory-plugins/blob/master/plugins/droid-evolved/skills/browser-navigation/SKILL.md

Start, stop, and restart video recording of browser sessions. Recordings are saved as WebM files.

```bash
agent-browser record start ./demo.webm
```

```bash
agent-browser click @e1
```

```bash
agent-browser record stop
```

```bash
agent-browser record restart ./take2.webm
```

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

### Two-Column Slide Layout Example

Source: https://github.com/factory-ai/factory-plugins/blob/master/plugins/droid-evolved/skills/visual-design/presentations.md

Demonstrates a two-column layout using '::right::' to separate content for each column.

```markdown
---
layout: two-cols
---

# Left Column

Content here

::right::

# Right Column

More content
```

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

### Vue Component Usage

Source: https://github.com/factory-ai/factory-plugins/blob/master/plugins/droid-evolved/skills/visual-design/reference-slide-example.md

Examples of embedding custom and built-in Vue components in slides.

```html
<Counter :count="10" />
```

```html
<Tweet id="1390115482657726468" />
```

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

### Example Experiment Grouping

Source: https://github.com/factory-ai/factory-plugins/blob/master/plugins/autoresearch/skills/autoresearch/SKILL.md

Illustrates how to group experiments into logical changesets, specifying files modified, metric improvements, and associated experiment IDs. This format is used for user presentation and approval.

```text
Group 1: "Reduce model depth from 8 to 6"
  Files: train.py (DEPTH, HEAD_DIM, N_EMBED)
  Metric improvement: val_bpb 1.15 -> 1.08 (-6.1%)
  Experiments: #3, #7, #12

Group 2: "Switch to cosine LR schedule"
  Files: train.py (lr_schedule, warmup_steps)
  Metric improvement: val_bpb 1.08 -> 1.05 (-2.8%)
  Experiments: #15, #18
```

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

### Enable Remote Access for Slidev

Source: https://github.com/factory-ai/factory-plugins/blob/master/plugins/droid-evolved/skills/visual-design/presentations.md

Starts the development server with remote access enabled, allowing control from other devices.

```bash
slidev --remote
```

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

### Threat Model Generation Invocations

Source: https://github.com/factory-ai/factory-plugins/blob/master/plugins/security-engineer/skills/threat-model-generation/SKILL.md

Example commands for generating or updating threat models based on project requirements.

```text
Generate a threat model for this repository using the threat-model-generation skill.
```

```text
Update the threat model - we added a new payments service in services/payments/.
```

```text
Generate a threat model for this repository. We need to comply with SOC2 and GDPR.
```

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

### Example PR Security Review Scan

Source: https://github.com/factory-ai/factory-plugins/blob/master/plugins/security-engineer/skills/security-review/SKILL.md

Use this command to scan a specific pull request for security vulnerabilities.

```bash
Scan PR #123 for security vulnerabilities.
```

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

### Define Slide Deck Structure

Source: https://github.com/factory-ai/factory-plugins/blob/master/plugins/droid-evolved/skills/visual-design/presentations.md

Example of a multi-slide deck using frontmatter, two-column layouts, and Mermaid diagrams.

```markdown
---
theme: seriph
background: https://cover.sli.dev
class: text-center
transition: slide-left
---

# Project Update

Q1 2026 Review

---

## layout: two-cols

# What we shipped

- Feature A
- Feature B
- Feature C

::right::

# Metrics

- 50% faster
- 2x more users
- 99.9% uptime

---

# Architecture

```mermaid
graph LR
  Client --> API
  API --> DB
  API --> Cache
```
```

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

### Video Recording API

Source: https://github.com/factory-ai/factory-plugins/blob/master/plugins/droid-evolved/skills/browser-navigation/SKILL.md

Commands to start, stop, and restart video recording of browser sessions.

```APIDOC
## Video Recording Commands

### Description
Commands to record browser sessions as video.

### Method
CLI Command

### Endpoint
N/A

### Parameters
#### Path Parameters
None

#### Query Parameters
None

#### Request Body
None

### Request Example
```bash
agent-browser record start ./demo.webm
agent-browser click @e1
agent-browser record stop
agent-browser record restart ./take2.webm
```

### Response
#### Success Response (200)
Video recording starts, stops, or restarts as specified.

#### Response Example
N/A
```

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

### Initialize Slidev Project

Source: https://github.com/factory-ai/factory-plugins/blob/master/plugins/droid-evolved/skills/visual-design/presentations.md

Use this command to create a new Slidev project from scratch.

```bash
npm init slidev@latest
```

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

### Manage presentations with Slidev

Source: https://github.com/factory-ai/factory-plugins/blob/master/plugins/droid-evolved/skills/visual-design/SKILL.md

Initialize a new project and use the CLI commands to run a development server, export to PowerPoint, or build for production.

```bash
npm init slidev@latest
slidev                    # dev server
slidev export --format pptx   # export to PowerPoint
slidev build              # build as hostable SPA
```

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

### Initialize Autoresearch Experiment

Source: https://context7.com/factory-ai/factory-plugins/llms.txt

Set up the experiment environment and baseline logging for optimization research.

```bash
# Create experiment branch and initialize state files
git checkout -b autoresearch/minimize-val-bpb-2025-01-15

# Initialize the experiment log
python3 autoresearch_helper.py init \
  --jsonl autoresearch.jsonl \
  --name 'minimize-val-bpb' \
  --metric-name 'val_bpb' \
  --direction lower

# Run baseline and log it
bash autoresearch.sh
python3 autoresearch_helper.py log \
  --jsonl autoresearch.jsonl \
  --commit $(git rev-parse --short=7 HEAD) \
  --metric 1.15 \
  --status keep \
  --description "baseline" \
  --asi '{"hypothesis": "baseline measurement"}'
```

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

### Initialize Experiment Log

Source: https://github.com/factory-ai/factory-plugins/blob/master/plugins/autoresearch/skills/autoresearch/SKILL.md

Use this command to initialize the experiment log file with details about the experiment's goal, metric name, and direction.

```bash
python3 autoresearch_helper.py init --jsonl autoresearch.jsonl --name '<goal>' --metric-name '<metric_name>' --direction <lower|higher>
```

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

### Build Slidev Presentation with PDF Download

Source: https://github.com/factory-ai/factory-plugins/blob/master/plugins/droid-evolved/skills/visual-design/presentations.md

Builds the SPA and includes a button for downloading the presentation as a PDF.

```bash
slidev build --download
```

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

### Basic Skill File Structure

Source: https://github.com/factory-ai/factory-plugins/blob/master/plugins/droid-evolved/skills/skill-creation/SKILL.md

Illustrates the directory structure for a custom skill, including the SKILL.md file.

```bash
.factory/skills/my-skill/
└── SKILL.md
```

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

### Upload Wiki Documentation

Source: https://context7.com/factory-ai/factory-plugins/llms.txt

Upload generated documentation to the Factory platform using the droid CLI.

```bash
# The wiki skill surveys the repository, plans a table of contents,
# generates pages with sub-agent delegation, and uploads to Factory

# Standard upload (keeps local wiki directory)
droid wiki-upload \
  --repo-url "$REPO_URL" \
  --wiki-dir ./droid-wiki

# Remote-only upload (deletes local files after upload)
droid wiki-upload \
  --repo-url "$REPO_URL" \
  --wiki-dir "$WIKI_TMPDIR" \
  --cleanup
```

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

### Code Review Suggestion Format

Source: https://context7.com/factory-ai/factory-plugins/llms.txt

Example of a code suggestion block provided by the review skill.

```markdown
## [P0] SQL Injection in user query handler

**File:** `src/api/users.js:45-49`

User input from `req.query.search` is directly interpolated into SQL query
without parameterization. Attacker can submit `test' OR '1'='1` to bypass
search filter and retrieve all users.

```suggestion
- const query = `SELECT * FROM users WHERE name LIKE '%${search}%'`;
- const results = await db.query(query);
+ const query = `SELECT * FROM users WHERE name LIKE $1`;
+ const results = await db.query(query, [`%${search}%`]);
```
```

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

### Natural language validation invocations

Source: https://github.com/factory-ai/factory-plugins/blob/master/plugins/security-engineer/skills/vulnerability-validation/SKILL.md

Example prompts for triggering security finding validation processes.

```text
Validate the security findings from the last scan.
```

```text
Validate findings VULN-001 and VULN-002 from security-findings.json.
```

```text
Validate all HIGH and CRITICAL severity findings from the security scan.
```

```text
Validate findings in security-findings.json using threat model at .factory/threat-model.md.
```

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

### Click Animation Directives

Source: https://github.com/factory-ai/factory-plugins/blob/master/plugins/droid-evolved/skills/visual-design/reference-slide-example.md

Examples of using v-click and v-mark directives for interactive slide elements.

```html
<div v-click>This shows up when you click the slide.</div>
```

```html
<span v-mark.underline.orange>inline markers</span>
```

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

### Build Commands

Source: https://github.com/factory-ai/factory-plugins/blob/master/plugins/droid-evolved/skills/visual-design/presentations.md

Commands for building the presentation into a static site.

```APIDOC
## Build Commands

### Description
Builds the presentation as a hostable Single Page Application (SPA).

### Endpoint
`slidev build`

### Parameters
#### Query Parameters
- **--out** (string) - Optional - Custom output directory (default: ./dist)
- **--download** (boolean) - Optional - Include PDF download button in SPA
- **--without-notes** (boolean) - Optional - Exclude speaker notes from the build
```

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

### Build Slidev Presentation to Custom Output Directory

Source: https://github.com/factory-ai/factory-plugins/blob/master/plugins/droid-evolved/skills/visual-design/presentations.md

Builds the presentation and places the output files in a specified directory.

```bash
slidev build --out ./public
```

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

### Slidev Magic Move Code Animation

Source: https://github.com/factory-ai/factory-plugins/blob/master/plugins/droid-evolved/skills/visual-design/presentations.md

Example of using 'magic-move' to animate code changes between steps.

```markdown
````md magic-move
```ts
// Step 1
const count = ref(0);
```

```ts
// Step 2
const count = ref(0);
const doubled = computed(() => count.value * 2);
```
````
```

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

### Create sequential stories

Source: https://github.com/factory-ai/factory-plugins/blob/master/plugins/droid-evolved/skills/visual-design/image-generation.md

Generate multi-step images for tutorials or process narratives.

```bash
nanobanana story "seed growing into tree" --steps=5 --type=process
nanobanana story "making coffee" --type=tutorial --steps=6
nanobanana story "day to night transition" --type=timeline --steps=4
```

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

### Frontend Design Guidelines

Source: https://context7.com/factory-ai/factory-plugins/llms.txt

Standardized design tokens and responsive breakpoints for web interface development.

```bash
# Spacing scale (use multiples of 4px or 8px)
# 4, 8, 12, 16, 24, 32, 48, 64, 96, 128

# Type scale (1.25 or 1.333 ratio)
# 12, 14, 16, 20, 24, 32, 40, 48

# Color palette (start minimal)
# - One primary color (brand/accent)
# - Neutrals: white, black, 3-4 grays
# - Semantic: red (errors), green (success)

# Responsive breakpoints
# Mobile: up to 640px
# Tablet: 641px to 1024px
# Desktop: 1025px and up
```

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

### List project folders

Source: https://github.com/factory-ai/factory-plugins/blob/master/plugins/core/skills/session-navigation/SKILL.md

Use these commands to view all project directories containing sessions or filter for a specific project.

```bash
# See all project folders with sessions
ls ~/.factory/sessions/

# Find folders for a specific project (partial match)
ls ~/.factory/sessions/ | grep "myapp"
```

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

### Build Slidev Presentation as SPA

Source: https://github.com/factory-ai/factory-plugins/blob/master/plugins/droid-evolved/skills/visual-design/presentations.md

Compiles the presentation into a static Single Page Application for hosting.

```bash
slidev build
```

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

### Log Baseline Result

Source: https://github.com/factory-ai/factory-plugins/blob/master/plugins/autoresearch/skills/autoresearch/SKILL.md

Parse the baseline metric from the output and log it as a 'keep' status in the JSONL file. This establishes the starting point for future comparisons.

```bash
python3 autoresearch_helper.py log --jsonl autoresearch.jsonl \
  --commit $(git rev-parse --short=7 HEAD) \
  --metric <baseline_value> \
  --status keep \
  --description "baseline" \
  --asi '{"hypothesis": "baseline measurement"}'
```

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

### Manage Browser Cookies and Storage

Source: https://github.com/factory-ai/factory-plugins/blob/master/plugins/droid-evolved/skills/browser-navigation/SKILL.md

Commands to get, set, and clear browser cookies and local storage. Useful for managing session data and user preferences.

```bash
agent-browser cookies                     # Get all cookies
```

```bash
agent-browser cookies set name value      # Set cookie
```

```bash
agent-browser cookies clear               # Clear cookies
```

```bash
agent-browser storage local               # Get all localStorage
```

```bash
agent-browser storage local key           # Get specific key
```

```bash
agent-browser storage local set k v       # Set value
```

```bash
agent-browser storage local clear         # Clear all
```

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

### Task Tool Invocation for Subagents

Source: https://github.com/factory-ai/factory-plugins/blob/master/plugins/core/skills/review/SKILL.md

Example of invoking a Task tool to spawn a 'file-group-reviewer' subagent for parallel code review. Ensure all subagents are spawned in a single response.

```json
{
  "subagent_type": "file-group-reviewer",
  "description": "Review auth module",
  "prompt": "Must include the PR context, the list of assigned files, the relevant diff sections, and instructions to return a JSON array of findings"
}
```

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

### Commit Initial State Files

Source: https://github.com/factory-ai/factory-plugins/blob/master/plugins/autoresearch/skills/autoresearch/SKILL.md

After initializing the experiment log, commit the relevant state files to version control.

```bash
git add autoresearch.md autoresearch.sh autoresearch.jsonl
git commit -m "autoresearch: initialize experiment session"
```

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

### Example JSON Output for Vulnerability VULN-002

Source: https://github.com/factory-ai/factory-plugins/blob/master/plugins/security-engineer/skills/vulnerability-validation/validation-examples.md

This JSON object represents the output for a vulnerability identified as VULN-002, indicating a FALSE_POSITIVE due to input validation by Joi schema middleware.

```json
{
  "id": "VULN-002",
  "original_severity": "HIGH",
  "status": "FALSE_POSITIVE",
  "reason": "Input is validated by Joi schema middleware before reaching this code. Schema enforces strict enum values ('electronics', 'clothing', 'food', 'other') which prevents any SQL injection payload from passing through.",
  "evidence": "See middleware/validation.js:23 - Joi.string().valid('electronics', 'clothing', 'food', 'other'). Middleware applied at routes/products.js:15 before handler."
}
```

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

### Run Baseline Benchmark

Source: https://github.com/factory-ai/factory-plugins/blob/master/plugins/autoresearch/skills/autoresearch/SKILL.md

Execute the main benchmark script to establish the baseline performance.

```bash
bash autoresearch.sh
```

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

### Create patterns and textures

Source: https://github.com/factory-ai/factory-plugins/blob/master/plugins/droid-evolved/skills/visual-design/image-generation.md

Generate seamless patterns and textures with configurable density and color schemes.

```bash
nanobanana pattern "hexagons" --style=geometric --colors=duotone
nanobanana pattern "marble texture" --type=texture
nanobanana pattern "subtle gradient" --type=wallpaper --density=sparse
```

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

### Theme Configuration

Source: https://github.com/factory-ai/factory-plugins/blob/master/plugins/droid-evolved/skills/visual-design/reference-slide-example.md

Frontmatter configuration for switching presentation themes.

```yaml
---
theme: default
---
```

```yaml
---
theme: seriph
---
```

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

### Skill Creation Structure

Source: https://context7.com/factory-ai/factory-plugins/llms.txt

File structure and YAML frontmatter format for creating new Droid skills.

```bash
# Skills live in a folder with a SKILL.md file
.factory/skills/my-skill/
└── SKILL.md

# Basic skill format with YAML frontmatter
```

```markdown
---
name: my-skill
version: 1.0.0
description: |
  What this skill does.
  When to use it.
---

# My skill

Instructions go here.
```

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

### Confirmed SQL Injection Validation

Source: https://github.com/factory-ai/factory-plugins/blob/master/plugins/security-engineer/skills/vulnerability-validation/validation-examples.md

Validates a confirmed SQL injection vulnerability by analyzing reachability, control flow, and exploitability. This example shows a direct string interpolation into an SQL query without sanitization.

```json
{
  "id": "VULN-001",
  "severity": "HIGH",
  "vulnerability_type": "SQL Injection",
  "file": "src/api/users.js",
  "line_range": "45-49",
  "code_snippet": "const sql = `SELECT * FROM users WHERE name LIKE '%${query}%'`;"
}
```

```json
{
  "payload": "test%' OR '1'='1' --",
  "request": "GET /api/users?search=test%25'%20OR%20'1'%3D'1'%20--",
  "expected_behavior": "Returns users with 'test' in name",
  "actual_behavior": "Returns ALL users in database"
}
```

```json
{
  "id": "VULN-001",
  "status": "CONFIRMED",
  "original_severity": "HIGH",
  "validated_severity": "CRITICAL",
  "exploitability": "EASY",
  "reachability": "EXTERNAL",
  "existing_mitigations": [],
  "exploitation_path": [
    "Attacker sends GET /api/users?search=<payload>",
    "Express passes query param to searchUsers()",
    "Function interpolates input into SQL template literal",
    "db.query() executes malicious SQL",
    "All user records returned to attacker"
  ],
  "proof_of_concept": {
    "payload": "test%' OR '1'='1' --",
    "request": "GET /api/users?search=test%25'%20OR%20'1'%3D'1'%20--",
    "expected_behavior": "Returns users with 'test' in name",
    "actual_behavior": "Returns ALL users in database"
  },
  "cvss_vector": "CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:N",
  "cvss_score": 9.1,
  "validation_notes": "Confirmed. Direct string interpolation in SQL with no input validation. Upgraded to CRITICAL due to ease of exploitation and data exposure."
}
```

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

### Standard Plugin Directory Structure

Source: https://context7.com/factory-ai/factory-plugins/llms.txt

The required file and folder layout for a valid Factory plugin.

```text
plugin-name/
├── .factory-plugin/
│   └── plugin.json       # Plugin metadata (name, description, author)
├── skills/               # Skill definitions
│   └── skill-name/
│       └── SKILL.md      # Skill instructions with YAML frontmatter
├── droids/               # Droid definitions (optional)
├── commands/             # Custom commands (optional)
├── mcp.json              # MCP server config (optional)
└── hooks.json            # Hook configurations (optional)
```

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

### Generate Visual Assets with Nanobanana

Source: https://context7.com/factory-ai/factory-plugins/llms.txt

Use the nanobanana CLI to generate images, icons, and diagrams, or manage presentations using Slidev.

```bash
# Install nanobanana CLI for image generation
npm install -g @factory/nanobanana
export GEMINI_API_KEY="your-key"

# Generate images
nanobanana generate "company logo" --count=4 --styles=modern,minimal
nanobanana edit photo.png "remove background"
nanobanana icon "settings gear" --style=flat
nanobanana diagram "auth flow" --type=flowchart

# Create presentations with Slidev
npm init slidev@latest
slidev                        # dev server
slidev export --format pptx   # export to PowerPoint
slidev build                  # build as hostable SPA
```

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

### Generate Assets with Nanobanana

Source: https://github.com/factory-ai/factory-plugins/blob/master/plugins/droid-evolved/skills/visual-design/presentations.md

Use the nanobanana CLI tool to generate diagrams, icons, and background patterns for slide content.

```bash
# Generate a diagram
nanobanana diagram "user authentication flow" --type=flowchart

# Generate an icon
nanobanana icon "rocket launch" --style=minimal

# Generate a background
nanobanana pattern "subtle gradient mesh" --type=wallpaper
```

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

### Full Repository Scan Command

Source: https://github.com/factory-ai/factory-plugins/blob/master/plugins/security-engineer/skills/security-review/SKILL.md

Initiate a comprehensive security scan across the entire repository.

```bash
@droid security --full
```

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

### Generate app icons

Source: https://github.com/factory-ai/factory-plugins/blob/master/plugins/droid-evolved/skills/visual-design/image-generation.md

Create icons with specific styles, types, and dimensions.

```bash
nanobanana icon "settings gear" --style=minimal
nanobanana icon "chat bubble" --type=favicon --sizes=16,32,64
nanobanana icon "rocket" --style=flat --background=transparent
```

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

### Identify projects by content

Source: https://github.com/factory-ai/factory-plugins/blob/master/plugins/core/skills/session-navigation/SKILL.md

Find which project directories contain sessions mentioning a specific keyword.

```bash
# Which projects have sessions mentioning "redis"?
rg -l "redis" ~/.factory/sessions/ | cut -d'/' -f1-5 | sort -u
```

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

### Interact with Web Elements using Refs

Source: https://github.com/factory-ai/factory-plugins/blob/master/plugins/droid-evolved/skills/browser-navigation/SKILL.md

Perform various interactions like clicking, typing, checking, and selecting elements using their references obtained from a snapshot. Supports single clicks, double clicks, hover, focus, text input, keyboard actions, form interactions, scrolling, drag-and-drop, and file uploads.

```bash
agent-browser click @e1
```

```bash
agent-browser dblclick @e1
```

```bash
agent-browser hover @e1
```

```bash
agent-browser focus @e1
```

```bash
agent-browser fill @e2 "text"
```

```bash
agent-browser type @e2 "text"
```

```bash
agent-browser press Enter
```

```bash
agent-browser press Control+a
```

```bash
agent-browser keydown Shift
```

```bash
agent-browser keyup Shift
```

```bash
agent-browser check @e1
```

```bash
agent-browser uncheck @e1
```

```bash
agent-browser select @e1 "value"
```

```bash
agent-browser scroll down 500
```

```bash
agent-browser scrollintoview @e1
```

```bash
agent-browser drag @e1 @e2
```

```bash
agent-browser upload @e1 file.pdf
```

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

### Basic Browser Navigation Commands

Source: https://github.com/factory-ai/factory-plugins/blob/master/plugins/droid-evolved/skills/browser-navigation/SKILL.md

Navigate websites, go back/forward, reload the page, or close the browser.

```bash
agent-browser open <url>
```

```bash
agent-browser back
```

```bash
agent-browser forward
```

```bash
agent-browser reload
```

```bash
agent-browser close
```

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

### Slidev Click-Through Code Highlighting

Source: https://github.com/factory-ai/factory-plugins/blob/master/plugins/droid-evolved/skills/visual-design/presentations.md

Shows how to use line highlighting syntax `{2|3-4|all}` for click-through animations in code blocks.

```markdown
```ts {2|3-4|all}
const greeting = "Hello";
const name = "World";
console.log(`${greeting}, ${name}!`);
```
```

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

### Skill Documentation - What Not To Do

Source: https://github.com/factory-ai/factory-plugins/blob/master/plugins/droid-evolved/skills/skill-creation/SKILL.md

Includes a section for documenting potential pitfalls or actions to avoid when using a skill.

```markdown
## What not to do

Don't run this on a dirty git working directory.
The `--force` flag will overwrite without asking.
```

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

### Authentication with Saved State

Source: https://github.com/factory-ai/factory-plugins/blob/master/plugins/droid-evolved/skills/browser-navigation/SKILL.md

Demonstrates logging into an application, saving the authentication state, and then loading that state in a subsequent session to bypass login.

```bash
# Login once
agent-browser open https://app.example.com/login
snapshots:
agent-browser snapshot -i
agent-browser fill @e1 "username"
agent-browser fill @e2 "password"
agent-browser click @e3
agent-browser wait --url "**/dashboard"
agent-browser state save auth.json

# Later sessions: load saved state
agent-browser state load auth.json
agent-browser open https://app.example.com/dashboard
```

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

### Generate Proof of Concept JSON

Source: https://github.com/factory-ai/factory-plugins/blob/master/plugins/security-engineer/skills/security-review/SKILL.md

Use this structure to document minimal PoCs for confirmed high or critical security findings.

```json
{
  "proof_of_concept": {
    "payload": "' OR '1'='1",
    "request": "GET /api/users?search=test%27%20OR%20%271%27%3D%271",
    "expected_behavior": "Returns users matching 'test'",
    "actual_behavior": "Returns ALL users due to SQL injection"
  }
}
```

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

### Manage Browser Tabs and Windows

Source: https://github.com/factory-ai/factory-plugins/blob/master/plugins/droid-evolved/skills/browser-navigation/SKILL.md

Commands for creating, switching, and closing browser tabs and windows. Useful for managing multiple pages or testing multi-window scenarios.

```bash
agent-browser tab                 # List tabs
```

```bash
agent-browser tab new [url]       # New tab
```

```bash
agent-browser tab 2               # Switch to tab
```

```bash
agent-browser tab close           # Close tab
```

```bash
agent-browser window new          # New window
```

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

### Export Slidev Presentation with Click Animations

Source: https://github.com/factory-ai/factory-plugins/blob/master/plugins/droid-evolved/skills/visual-design/presentations.md

Exports the presentation where each click animation results in a new page in the output.

```bash
slidev export --with-clicks
```

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

### Read session details

Source: https://github.com/factory-ai/factory-plugins/blob/master/plugins/core/skills/session-navigation/SKILL.md

Inspect session metadata, settings, and conversation length using standard command-line tools.

```bash
# The metadata (title, working directory)
head -1 ~/.factory/sessions/-Users-enoreyes-code-work-myapp/<uuid>.jsonl | jq .

# Session stats (model, tokens, duration)
cat ~/.factory/sessions/-Users-enoreyes-code-work-myapp/<uuid>.settings.json | jq .

# How long was this conversation?
wc -l ~/.factory/sessions/-Users-enoreyes-code-work-myapp/<uuid>.jsonl
```

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

### Run Experiment Benchmark with Timeout

Source: https://github.com/factory-ai/factory-plugins/blob/master/plugins/autoresearch/skills/autoresearch/SKILL.md

Execute the benchmark script for an experiment, with a specified timeout to prevent indefinite runs.

```bash
timeout 600 bash autoresearch.sh
```

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

### Factory Plugin Structure

Source: https://github.com/factory-ai/factory-plugins/blob/master/README.md

This outlines the standard directory structure for a Factory plugin. Key components include metadata in .factory-plugin/plugin.json, skill definitions in skills/, and optional droids/, commands/, mcp.json, and hooks.json.

```plaintext
plugin-name/
├── .factory-plugin/
│   └── plugin.json       # Plugin metadata
├── skills/               # Skill definitions
│   └── skill-name/
│       └── SKILL.md
├── droids/               # Droid definitions (optional)
├── commands/             # Custom commands (optional)
├── mcp.json              # MCP server config (optional)
└── hooks.json            # Hook configurations (optional)
```

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

### Run Experiment Loop

Source: https://context7.com/factory-ai/factory-plugins/llms.txt

Execute the autonomous experiment loop with a timeout.

```bash
# The experiment loop runs autonomously:
# 1. Choose what to try (read autoresearch.md for context)
# 2. Make focused changes (one hypothesis per experiment)
# 3. Run the experiment
timeout 600 bash autoresearch.sh
```

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

### Create Clean Git Branches for Experiment Groups

Source: https://github.com/factory-ai/factory-plugins/blob/master/plugins/autoresearch/skills/autoresearch/SKILL.md

Shell commands to create a new Git branch for each experiment group, checking out specific files from the session branch and committing with a detailed description including metric improvements and experiment counts.

```bash
merge_base=$(git merge-base HEAD main)
git checkout -b autoresearch/finalize/<group-name> $merge_base
git checkout autoresearch/<session-branch> -- <file1> <file2> ...
git commit -m "<group description>

Autoresearch results:
- Metric: <name> improved from <baseline> to <best> (<delta>%) 
- Confidence: <score>x noise floor
- Experiments: <count> total, <kept> kept"
```

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

### Generate Threat Models

Source: https://context7.com/factory-ai/factory-plugins/llms.txt

Commands and natural language prompts to generate and update repository threat models.

```bash
# Generate initial threat model
# Creates .factory/threat-model.md and .factory/security-config.json

# Invoke via natural language:
"Generate a threat model for this repository using the threat-model-generation skill."

# Update after architecture changes:
"Update the threat model - we added a new payments service in services/payments/."

# With compliance requirements:
"Generate a threat model for this repository. We need to comply with SOC2 and GDPR."
```