Instruction file imported from LastBotInc/cursor-rules-tools (.cursor/rules/command_line_tools.mdc). Copyright stays with the author.

<command_line_tools>

Project-Specific Tools: Execute these using run_terminal_cmd.

🔧 Setup: Run npm install and configure API keys in .env.local (see .env.example)

{ "tools": { "image-optimizer": { "description": "Optimizes images with background removal, resizing, and format conversion using Sharp and Replicate's remove-bg model", "tool": "run_terminal_cmd", "command": "npm run optimize-image", "status": "✅ Tested and working", "options": { "input": "Path to input image", "output": "Path to output image", "remove-bg": "(Optional) Remove image background using AI", "resize": "(Optional) Resize image (format: WIDTHxHEIGHT, e.g. 800x600)", "format": "(Optional) Convert to format (png, jpeg, or webp)", "quality": "(Optional) Set output quality (1-100, default: 80)" }, "requires": [ "REPLICATE_API_TOKEN in .env.local (for background removal)", ], "example": "npm run optimize-image -- -i input.png -o output.webp --resize 512x512 --format webp --quality 90", "note": "Use -i and -o flags for input/output. Background removal requires Replicate API token." }, "html-to-md": { "description": "Scrapes a webpage and converts its HTML content to Markdown format using Turndown service", "tool": "run_terminal_cmd", "command": "npm run html-to-md", "status": "✅ Tested and working", "options": { "url": "URL of the webpage to scrape", "output": "(Optional) Output file path for the markdown (default: output.md)", "selector": "(Optional) CSS selector to target specific content" }, "requires": [ ], "example": "npm run html-to-md -- --url https://example.com --output docs/scraped.md --selector main" }, "gemini": { "description": "Interacts with Google's Gemini API for text generation, chat, multimodal tasks, document analysis, and grounded search", "tool": "run_terminal_cmd", "command": "npm run gemini", "status": "✅ Tested and working", "options": { "prompt": "Text prompt or question for the model", "model": "(Optional) Model to use: 'gemini-2.0-flash' (default), 'gemini-2.5-pro-exp-03-25'", "temperature": "(Optional) Sampling temperature between 0.0 and 1.0 (default: 0.7)", "max-tokens": "(Optional) Maximum tokens to generate (default: 2048)", "image": "(Optional) Path to image file for vision tasks", "file": "(Optional) Path to local file (PDF, DOCX, TXT, etc.) for document analysis", "url": "(Optional) URL to a document to analyze (PDF, DOCX, TXT, etc.)", "mime-type": "(Optional) MIME type of the file (e.g., application/pdf, default: auto-detected)", "chat-history": "(Optional) Path to JSON file containing chat history", "stream": "(Optional) Stream the response (default: false)", "safety-settings": "(Optional) JSON string of safety threshold configurations", "schema": "(Optional) JSON schema for structured output", "json": "(Optional) Return structured JSON data. Available types: recipes, tasks, products, custom", "ground": "(Optional) Enable Google Search grounding for up-to-date information (default: false)", "show-search-data": "(Optional) Show the search entries used for grounding (default: false)" }, "requires": [ "GOOGLE_AI_STUDIO_KEY or GEMINI_API_KEY in .env.local", "@google/generative-ai package (auto-installed with npm install)", "node-fetch package (auto-installed with npm install)" ], "example": "npm run gemini -- --prompt "What is the capital of France?" --model gemini-2.0-flash --temperature 0.7", "advanced_examples": [ "# Process a PDF document from a URL", "npm run gemini -- --prompt "Summarize this document in 5 key points" --url "https://discovery.ucl.ac.uk/id/eprint/10089234/1/343019_3_art_0_py4t4l_convrt.pdf"", "", "# Process a local PDF file", "npm run gemini -- --prompt "What is this document about?" --file test/data/sample.pdf", "", "# Process a text file with specific MIME type", "npm run gemini -- --prompt "Expand on this information" --file test/data/sample.txt --mime-type text/plain", "", "# Get grounded search results with real-time information", "npm run gemini -- --prompt "When is the next total solar eclipse in North America?" --ground --show-search-data", "", "# Generate structured JSON data with predefined schema (recipes)", "npm run gemini -- --prompt "List 3 popular cookie recipes" --json recipes", "", "# Generate structured JSON data with predefined schema (tasks)", "npm run gemini -- --prompt "Create a task list for planning a vacation" --json tasks", "", "# Generate structured JSON data with predefined schema (products)", "npm run gemini -- --prompt "List 3 smartphone models" --json products", "", "# Use a custom JSON schema for structured output", "npm run gemini -- --prompt "List 3 programming languages and their use cases" --json custom --schema '{"type":"array","items":{"type":"object","properties":{"language":{"type":"string"},"year":{"type":"integer"},"useCases":{"type":"array","items":{"type":"string"}}},"required":["language","useCases"]}}'" ] }, "gemini-image": { "description": "Generates and edits images using Google Gemini and Imagen models", "tool": "run_terminal_cmd", "command": "node tools/gemini-image-tool.js", "status": "✅ Tested and working", "subcommands": { "generate": { "description": "Generate an image using Gemini 2.0 or Imagen 3.0", "options": { "prompt": "Required: Text prompt for image generation", "model": "(Optional) Model to use: 'gemini-2.0' (default) or 'imagen-3.0'", "output": "(Optional) Output file path (default: gemini-generated-image.png)", "folder": "(Optional) Output folder path (default: public/images)", "num-outputs": "(Optional) Number of images (Imagen 3 only, default: 1, max: 4)", "negative-prompt": "(Optional) Negative prompt (Imagen 3 only)", "aspect-ratio": "(Optional) Aspect ratio (Imagen 3 only, default: 1:1, options: 1:1, 16:9, 9:16, 4:3, 3:4)" }, "example": "node tools/gemini-image-tool.js generate -p "A futuristic car" -m imagen-3.0 -n 2 --aspect-ratio 16:9 -o car.png" }, "edit": { "description": "Edit an existing image using Gemini 2.0", "options": { "input-image": "Required: Path to the input image", "edit-prompt": "Required: Text prompt describing the edit", "output": "(Optional) Output file path (default: gemini-edited-image.png)", "folder": "(Optional) Output folder path (default: public/images)" }, "example": "node tools/gemini-image-tool.js edit -i input.png -p "Add sunglasses to the person" -o edited.png" } }, "requires": [ "GOOGLE_AI_STUDIO_KEY or GEMINI_API_KEY in .env.local", "google/genai package (auto-installed with npm install)", ] }, "download-file": { "description": "Downloads files from URLs with progress tracking, automatic file type detection, and customizable output paths", "tool": "run_terminal_cmd", "command": "npm run download-file", "status": "✅ Tested and working", "options": { "url": "URL of the file to download", "output": "(Optional) Complete output path including filename", "folder": "(Optional) Output folder path (default: downloads)", "filename": "(Optional) Output filename (if not provided, derived from URL or content)" }, "requires": [ ], "example": "npm run download-file -- --url https://example.com/image.jpg --folder public/images --filename downloaded-image.jpg" }, "generate-video": { "description": "Generates videos using various Replicate API models. Can optionally first generate an image using OpenAI (GPT-image-1) based on an image prompt, then use that image for video generation.", "tool": "run_terminal_cmd", "command": "npm run generate-video", "status": "⚠️ Requires API keys - needs REPLICATE_API_TOKEN configured", "options": { "prompt": "Text description of the desired video (used by Replicate).", "model": "(Optional) Replicate model to use: kling-1.6 (default), kling-2.0, minimax, hunyuan, mochi, or ltx.", "duration": "(Optional) Duration of the video in seconds (model-specific limits apply).", "image": "(Optional) Path to an existing input image for image-to-video generation. If using --image-prompt, this will typically be the same path as --openai-image-output.", "output": "(Optional) Output filename for the video.", "folder": "(Optional) Output folder path for the video (default: public/videos).", "image-prompt": "(Optional) Text prompt for OpenAI (GPT-image-1) to generate an initial image. This image will then be used for video generation.", "openai-image-output": "(Optional) Output path for the image generated by OpenAI (e.g., public/images/my-image.png). Required if --image-prompt is used and you want to specify the image name/path for the video step.", "aspect-ratio": "(Optional) Aspect ratio for the video (e.g., 16:9, 1:1). Support depends on the Replicate model." }, "requires": [ "REPLICATE_API_TOKEN in .env.local", "OPENAI_API_KEY in .env.local (if using --image-prompt)" ], "example": "npm run generate-video -- --image-prompt "A futuristic robot playing chess" --openai-image-output public/images/robot-chess.png --prompt "Animate the robot making a move, cinematic style" --image public/images/robot-chess.png --model kling-1.6 --duration 4 --output robot-chess-video.mp4" }, "remove-background-advanced": { "description": "Advanced background removal tool using Sharp with color tolerance and edge detection", "tool": "run_terminal_cmd", "command": "npm run remove-background-advanced", "status": "✅ Tested and working", "options": { "input": "Path to input image", "output": "Path to output image", "tolerance": "(Optional) Color tolerance for background detection (0-255, default: 30)" }, "requires": [], "example": "npm run remove-background-advanced -- --input input.png --output output.png --tolerance 40" }, "openai-image": { "description": "Generate and edit images using OpenAI's GPT-image-1 and DALL-E models", "tool": "run_terminal_cmd", "command": "npm run openai-image", "status": "✅ Tested and working", "subcommands": { "generate": { "description": "Generate an image using OpenAI's image generation models", "options": { "prompt": "Required: Text prompt for image generation", "model": "(Optional) Model to use: "gpt-image-1" or "dall-e-3" (default: gpt-image-1)", "output": "(Optional) Output file path (default: openai-generated-image.png)", "folder": "(Optional) Output folder path (default: public/images)", "size": "(Optional) Image size: 1024x1024, 1792x1024, or 1024x1792 (default: 1024x1024)", "quality": "(Optional) Image quality: standard or hd - DALL-E only (default: standard)", "number": "(Optional) Number of images to generate (1-4) - DALL-E only (default: 1)", "reference-image": "(Optional) Reference image path for gpt-image-1 with reference", "creative": "(Optional) Creativity level for GPT-image-1: standard or vivid (default: standard)" }, "example": "npm run openai-image -- generate -p "A futuristic cityscape at sunset with flying cars" -m gpt-image-1 -s 1792x1024 -c vivid" }, "edit": { "description": "Edit an existing image using OpenAI's models", "options": { "input-image": "Required: Path to the input image to edit", "edit-prompt": "Required: Text prompt describing the edit to apply", "model": "(Optional) Model to use: "gpt-image-1" or "dall-e-3" (default: gpt-image-1)", "output": "(Optional) Output file path (default: openai-edited-image.png)", "folder": "(Optional) Output folder path (default: public/images)", "size": "(Optional) Image size: 1024x1024, 1792x1024, or 1024x1792 (default: 1024x1024)", "creative": "(Optional) Creativity level for GPT-image-1: standard or vivid (default: standard)" }, "example": "npm run openai-image -- edit -i input.jpg -p "Change the background to a tropical beach" -m gpt-image-1" } }, "requires": [ "OPENAI_API_KEY in .env.local" ] }, "github-cli": { "description": "Interact with GitHub repositories, pull requests, issues, and workflows directly from the command line", "tool": "run_terminal_cmd", "command": "npm run github", "status": "🔧 Not tested - requires GitHub authentication", "options": { "auth": { "description": "Manage GitHub authentication", "options": { "login": "(Optional) Log in to GitHub (-l, --login)", "status": "(Optional) View authentication status (-s, --status)", "refresh": "(Optional) Refresh stored authentication credentials (-r, --refresh)", "logout": "(Optional) Log out of GitHub (-o, --logout)" } }, "pr-create": { "description": "Create a pull request", "options": { "title": "(Optional) Pull request title", "body": "(Optional) Pull request body", "base": "(Optional) Base branch name", "draft": "(Optional) Create draft pull request", "fill": "(Optional) Use commit info for title/body" } }, "pr-list": { "description": "List and filter pull requests", "options": { "state": "(Optional) Filter by state: open, closed, merged, all (default: open)", "limit": "(Optional) Maximum number of items to fetch (default: 30)", "assignee": "(Optional) Filter by assignee", "author": "(Optional) Filter by author", "base": "(Optional) Filter by base branch", "web": "(Optional) Open list in the browser" } }, "pr-view": { "description": "View pull request details", "arguments": "[number]: PR number or URL, if omitted the PR from the current branch is displayed", "options": { "web": "(Optional) Open in web browser" } }, "issue-create": { "description": "Create a new issue", "options": { "title": "(Optional) Issue title", "body": "(Optional) Issue body", "assignee": "(Optional) Assign people by their login", "label": "(Optional) Add labels by name", "project": "(Optional) Add issue to project", "milestone": "(Optional) Add issue to milestone", "web": "(Optional) Open new issue in the web browser" } }, "issue-list": { "description": "List and filter issues", "options": { "state": "(Optional) Filter by state: open, closed, all (default: open)", "limit": "(Optional) Maximum number of issues to fetch (default: 30)", "assignee": "(Optional) Filter by assignee", "author": "(Optional) Filter by author", "label": "(Optional) Filter by label", "milestone": "(Optional) Filter by milestone", "web": "(Optional) Open list in the browser" } }, "release-create": { "description": "Create a new release", "arguments": ": Tag name for the release", "options": { "title": "(Optional) Release title", "notes": "(Optional) Release notes", "notes-file": "(Optional) Read release notes from file", "draft": "(Optional) Save release as draft instead of publishing", "prerelease": "(Optional) Mark release as prerelease", "generate-notes": "(Optional) Automatically generate release notes" } }, "release-list": { "description": "List releases in a repository", "options": { "limit": "(Optional) Maximum number of releases to fetch (default: 30)" } }, "repo": { "description": "Manage repositories", "options": { "create": "(Optional) Create a new repository", "description": "(Optional) Repository description", "homepage": "(Optional) Repository homepage URL", "private": "(Optional) Make the repository private", "team": "(Optional) The name of the organization team to grant access to", "view": "(Optional) View repository details", "web": "(Optional) Open in web browser", "list": "(Optional) List your repositories" } }, "workflow": { "description": "Manage GitHub workflows", "options": { "list": "(Optional) List workflows", "run": "(Optional) Run a workflow by name or ID", "view": "(Optional) View a specific workflow", "enable": "(Optional) Enable a workflow by name or ID", "disable": "(Optional) Disable a workflow by name or ID" } }, "tasks": { "description": "List and manage tasks from GitHub Projects and Issues", "options": { "project": "(Optional) Specify project ID or number to list tasks from", "repository": "(Optional) Specify repository in owner/repo format (default: current repository)", "status": "(Optional) Filter by status: open, closed, all (default: open)", "label": "(Optional) Filter by label", "assignee": "(Optional) Filter by assignee", "limit": "(Optional) Maximum number of tasks to fetch (default: 30)", "format": "(Optional) Output format: table, json, or simple (default: table)", "web": "(Optional) Open tasks in the web browser" } } }, "requires": [ "GitHub CLI (will be auto-installed if not present)", "GitHub account authentication" ], "example": "npm run github -- pr-create --title "New feature" --body "This PR adds a new feature"" } } } </command_line_tools>