// MCP TOOL REFERENCE

SPARKFORGE TOOLS

33 tools from the SparkForge MCP Server, categorised by risk level.

View the SparkForge policy → Token cost: 2,716 tokens →

// ALL 33 SPARKFORGE TOOLS

READ 25 tools

Read btc_signal Get a BTC price direction prediction based on 10-indicator technical analysis consensus. Returns up/down/ne... Read check_domain Check domain availability and DNS info via lookup. Read code_review Review code using Claude AI. Returns bugs, improvements, and best practice suggestions. Read crypto_news_summary Get latest crypto news headlines from CryptoPanic. Read detect_language Detect the language of a text string. Supports 10+ languages. Read extract_structured Extract structured JSON data from a URL using Claude AI and a provided JSON schema. Read get_crypto_price Get current cryptocurrency price, 24h change, and market cap from CoinGecko. Read get_url_metadata Extract metadata from a URL — title, description, Open Graph tags, favicon, canonical URL. Read html_to_markdown Convert HTML content or fetch a URL and convert its HTML to clean Markdown. Read list_services List all available SparkForge services with pricing, parameters, and demo URLs. Read ocr Extract text from an image URL using Claude vision AI. Read pdf_to_markdown Extract text from a PDF URL and convert to Markdown. Returns the text content with page count. Read scrape_webpage Scrape a webpage and return clean readable text/markdown. Strips scripts, styles, and HTML tags. Read site_crawl Crawl a website starting from a URL, follow internal links, return pages as text. Read smart_summarize AI-powered summarization in 4 modes: bullet_points, executive, technical, eli5. Read store_data Key-value data storage. Operations: set, get, delete, list. Data persists per namespace. Read summarize_text Summarize long text using extractive summarization. Fast, algorithmic — no AI API call needed. Read take_screenshot Capture a screenshot of any webpage URL. Returns the screenshot as a base64-encoded PNG image. Read text_diff Compare two texts and show a unified diff with additions and removals. Read text_to_speech Convert text to speech using ElevenLabs. Returns audio as base64 MP3. Read token_research Research a crypto token — description, links, price, market cap, volume from CoinGecko. Read validate_json Validate and prettify JSON. Optionally validate against a JSON Schema. Read verify_email Validate email format and check MX records for the domain. Read wallet_analysis Analyze an Ethereum/Base wallet — ETH balance and recent transactions via Etherscan. Read web_search Search the web using Brave Search API. Returns titles, URLs, and descriptions.

WRITE 8 tools

Write convert_format Convert between formats: csv→json, json→csv, markdown→html, html→text. Write generate_image Generate an AI image from a text prompt using Flux models. Returns an image URL. Models: flux-schnell (fast... Write generate_meme Generate a meme image using Imgflip API. Returns a meme image URL. Write generate_qr_code Generate a QR code as a base64-encoded PNG image from any text or URL. Write generate_video Generate a short AI video from a text prompt using fal.ai. Write image_resize Fetch an image from a URL, resize it, and return as base64 PNG. Alias for resize_image. Write resize_image Fetch an image from a URL, resize it, and return as base64 PNG. Write write_tweet Generate tweet text from a topic or prompt using Claude AI. Returns tweet-ready text under 280 chars.

// FAQ

How many tools does the SparkForge MCP server have? +

The SparkForge MCP server exposes 33 tools across 2 categories: Read, Write.

How do I enforce policies on SparkForge tools? +

Route the SparkForge server through the PolicyLayer gateway. Define allow, deny, or approval rules per tool in the dashboard — they are enforced on every call before it reaches the server.

What risk categories do SparkForge tools fall into? +

SparkForge tools are categorised as Read (25), Write (8). Each category has a recommended default policy.

Let agents act without letting them run wild.

Route your MCP servers through PolicyLayer and every tool call is checked against your policy before it runs — allow, deny, or require approval. Per-identity grants. Full audit log. Live in minutes.

SECURE YOUR MCP →

Free to start. No card required.

4,600+ MCP servers and 31,000+ tools scanned and risk-classified.