Generate short-form video script text (hook/educational/cta styles).

Get Lead by LinkedIn URL

Update a brand's name, regex, aliases, or domains. Changes to name/regex/aliases trigger background metric recalculation; repeat attempts during recalculation will fail. Confirm with the user before calling.

Semantic search across all extracted datasheets. Finds components matching natural language queries about specifications, features, or capabilities. Best for broad spec-based discovery across all parts (e.g. 'low-noise LDO with PSRR above 70dB'). Only searches datasheets that have been previously extracted — not all parts that exist. For finding specific parts by number, use search_parts instead.

Create a manual backup (runs asynchronously). The backup starts in the background. Poll list_backups() to check status. Requires: API key with write scope. Args: slug: Site identifier Returns: {"id": "uuid", "status": "pending", "message": "Backup started. Poll list_backups() to check status."}

Search across ALL string properties of ALL nodes in a deployed graph using free-text queries. Unlike search_graph_nodes (which filters by specific property), this searches every text field at once. Perfect for finding knowledge when you don't know which property contains the answer. Example: query "quantum" searches name, description, summary, notes, and all other string fields. Returns nodes with _match_fields showing which properties matched. Optionally filter by entity_type to narrow results.

Execute raw, client-provided SQL queries against an ephemeral database initialized with the provided schema. Returns query results in a simple JSON format with column headers and row data as a 2D array. The database type (SQLite or Postgres) is specified via the databaseType parameter: - SQLITE: In-memory, lightweight, uses standard SQLite syntax - POSTGRES: Temporary isolated schema with dedicated user, uses PostgreSQL syntax and features WHEN TO USE: When you need to run your own hand-written SQL queries to test database behavior or compare the output with ExoQuery results from validateAndRunExoquery. This lets you verify that ExoQuery-generated SQL produces the same results as your expected SQL. INPUT REQUIREMENTS: - query: A valid SQL query (SELECT, INSERT, UPDATE, DELETE, etc.) - schema: SQL schema with CREATE TABLE and INSERT statements to initialize the test database - databaseType: Either "SQLITE" or "POSTGRES" (defaults to SQLITE if not specified) OUTPUT FORMAT: On success, returns JSON with the SQL query and a 2D array of results: {"sql":"SELECT * FROM users ORDER BY id","output":[["id","name","age"],["1","Alice","30"],["2","Bob","25"],["3","Charlie","35"]]} Output format details: - First array element contains column headers - Subsequent array elements contain row data - All values are returned as strings On error, returns JSON with error message and the attempted query (if available): {"error":"Query execution failed: no such table: USERS","sql":"SELECT * FROM USERS"} Or if schema initialization fails: {"error":"Database initialization failed due to: near \"CREAT\": syntax error\\nWhen executing the following statement:\\n--------\\nCREAT TABLE users ...\\n--------","sql":"CREAT TABLE users ..."} EXAMPLE INPUT: Query: SELECT * FROM users ORDER BY id Schema: CREATE TABLE users ( id INTEGER PRIMARY KEY, name TEXT NOT NULL, age INTEGER ); INSERT INTO users (id, name, age) VALUES (1, 'Alice', 30); INSERT INTO users (id, name, age) VALUES (2, 'Bob', 25); INSERT INTO users (id, name, age) VALUES (3, 'Charlie', 35); EXAMPLE SUCCESS OUTPUT: {"sql":"SELECT * FROM users ORDER BY id","output":[["id","name","age"],["1","Alice","30"],["2","Bob","25"],["3","Charlie","35"]]} EXAMPLE ERROR OUTPUT (bad table name): {"error":"Query execution failed: no such table: invalid_table","sql":"SELECT * FROM invalid_table"} EXAMPLE ERROR OUTPUT (bad schema): {"error":"Database initialization failed due to: near \"CREAT\": syntax error\\nWhen executing the following statement:\\n--------\\nCREAT TABLE users (id INTEGER)\\n--------\\nCheck that the initialization SQL is valid and compatible with SQLite.","sql":"CREAT TABLE users (id INTEGER)"} COMMON QUERY EXAMPLES: Select all rows: SELECT * FROM users Select specific columns with filtering: SELECT name, age FROM users WHERE age > 25 Aggregate functions: SELECT COUNT(*) as total FROM users Join queries: SELECT u.name, o.total FROM users u JOIN orders o ON u.id = o.user_id Insert data: INSERT INTO users (name, age) VALUES ('David', 40) Update data: UPDATE users SET age = 31 WHERE name = 'Alice' Delete data: DELETE FROM users WHERE age < 25 Count with grouping: SELECT age, COUNT(*) as count FROM users GROUP BY age SCHEMA RULES: - Use standard SQLite syntax - Table names are case-sensitive (use lowercase for simplicity or quote names) - Include INSERT statements to populate test data for meaningful results - Supported data types: INTEGER, TEXT, REAL, BLOB, NULL - Use INTEGER PRIMARY KEY for auto-increment columns - Schema SQL is split on semicolons (;), so each statement after a ';' is executed separately - Avoid semicolons in comments as they will cause statement parsing issues COMPARISON WITH EXOQUERY: This tool is designed to work alongside validateAndRunExoquery for comparison purposes: 1. Use validateAndRunExoquery to run ExoQuery Kotlin code and see the generated SQL + results 2. Use runRawSql with your own hand-written SQL to verify you get the same output 3. Compare the outputs to ensure ExoQuery generates the SQL you expect 4. Test edge cases with plain SQL before writing equivalent ExoQuery code

Execute a saved Workflow on one or more images. Runs a previously created Workflow against the provided images on the Roboflow serverless infrastructure. IMPORTANT: If processing more than 10 images, spawn a sub-agent to run this tool in the background so the user is not blocked. Returns workflow outputs as defined by the workflow's output blocks.

Get the builder workflows — step-by-step state machines for building skills and solutions. Use this to guide users through the entire build process conversationally. Returns phases, what to ask, what to build, exit criteria, and tips for each stage.

Search the web for any topic and get clean, ready-to-use content. Best for: Finding current information, news, facts, people, companies, or answering questions about any topic. Returns: Clean text content from top search results. Query tips: describe the ideal page, not keywords. "blog post comparing React and Vue performance" not "React vs Vue". Use category:people / category:company to search through Linkedin profiles / companies respectively. If highlights are insufficient, follow up with web_fetch_exa on the best URLs.

Start a contact research job — AI gathers insights from LinkedIn and other sources asynchronously; use contacts.get_research to poll for results

Recommend the right DezignWorks product tier based on the user's equipment and needs. DezignWorks offers three tiers: Probing ($6,995 — for FARO/Romer portable CMM users), Mesh Modeler ($8,995 — for handheld 3D scanner users), and Unlimited ($12,995 — for users who need both probing and scanning). Use this tool when an agent needs to match a customer's hardware or workflow to the correct product.

Discover available DOOH screens in the Trillboards network. WHEN TO USE: - Finding screens by venue type (retail, transit, office, etc.) - Finding screens in a specific city/state or within a radius - Finding screens with a specific audience profile (high income, professionals, etc.) - Getting an overview of available inventory with live audience data RETURNS: - screens: Array of screen objects with location, venue type, online status, and live audience data - total: Total matching screens - online_count: Number of currently online screens Each screen includes real-time audience data when available: - face_count, attention_score, income_level, mood, lifestyle - purchase_intent, crowd_density, ad_receptivity, dwell_time EXAMPLE: User: "Find retail screens in New York with high-income audience" discover_inventory({ venue_types: ["retail"], location: { city: "New York", state: "NY" }, audience_profile: { income: "high" }, limit: 20 })

Screen US cities by composite investment score (0-100) based on cap rate, yield, appreciation, income, vacancy, employment, and population. Pre-computed for 18,800+ cities. Ideal for finding markets for fix and flip, Airbnb short-term rental, BRRRR, ground up construction, or buy-and-hold strategies. USDV Capital — Your Real Estate CFO.

Search specific CFR violation citations from FDA inspections (Compliance Dashboard data, not available in openFDA API). Filter by company name, FEI number, CFR number (e.g., '21 CFR 211.68' for a specific section, or '21 CFR 211' for all cGMP violations), or keyword in citation descriptions. Returns the cited regulation, short and long descriptions of the finding, and inspection dates. Related: fda_inspections (inspection classification and dates by FEI), fda_compliance_actions (warning letters that may reference these citations).

Generate and plot synthetic financial price data (requires matplotlib). Creates realistic price movement patterns for educational purposes. Does not use real market data. Note: Use for time-series price data with optional moving average overlay. For general XY data, use plot_line_chart instead. Examples: plot_financial_line(days=60, trend='bullish') plot_financial_line(days=90, trend='volatile', start_price=150.0, color='orange')

Re-check a specific control after applying a fix. Confirms whether the finding is resolved.

Generate a complete 3D scene from a structured scene plan. Your job: - Convert the scene plan into structured scene data - Drive materials, lighting, background, and layout from design_tokens when present - Apply user-specified color hints to background and accent colors - Propagate design_tokens through to scene_data.metadata for downstream tools Rules: - Do NOT modify the scene plan - Do NOT add new objects - Use provided objects exactly - First object is the main subject - Apply style and animation as given - Consume design_tokens directly when present This tool is deterministic and does not interpret intent.

Remove image background using AI (U2-Net). Returns RGBA PNG/WebP with transparent background. Perfect for product photos, portraits, and design assets. FREE. (FREE)

Analyze non-email messages (SMS, WhatsApp, Instagram DMs, Discord, Slack, Telegram, LinkedIn, Facebook Messenger, iMessage, Signal) for platform-specific threats including smishing, wrong-number scams, OTP interception, impersonation, and crypto fraud. $0.01/call via skyfire-api-key header (Skyfire Buyer API Key). By using this service you accept the Terms of Service. Advisory service only.

Search for land parcels by parcel ID prefix (autocomplete). Returns matching parcels with their district, area, and GPS coordinates. Useful for finding exact parcel IDs, then searching transactions nearby. Example: search for parcels starting with '146518_8.01'.

List lead lists (batch jobs) in the user's MachFive workspace. Useful for browsing past batches, checking what's in progress, or finding a list_id to export. Results are ordered newest first.

Given a product ID, find similar products across the entire catalog. Useful for "more like this" recommendations or finding alternatives. Returns up to 10 results per page, paginated (max 3 pages).

List all available legal agreement templates. Supports compact and full metadata modes for browsing. For finding templates by topic, jurisdiction, or source, use search_templates instead.

Search saved contexts by keyword or natural language query. Returns matching context summaries ranked by relevance using hybrid keyword + semantic search. Searches across titles, reasoning, code snippets, file paths, and commit SHAs. Results include local contexts and (if authenticated) cloud workspace contexts including team-shared contexts. Use for finding past decisions, understanding why code was written, or discovering relevant prior work. Use gitwhy_get <context_id> to read the full context. CLI alternative: `git why search "<query>"`.

Search MetaMask Embedded Wallets (Web3Auth) documentation and examples. Use for SDK discovery, feature lookup, and finding relevant examples. Returns doc page links with snippets and matching example projects.

Scan up to 10 domains at once. Returns score, grade, and finding counts per domain.

Generate a pentest report finding in Markdown format for a specific vulnerability. Fetches full detail and formats it as a professional finding with severity, CVSS, description, affected products, exploit availability, and references. Accepts both CVE-IDs and EIP-IDs. Optionally include the target system tested and tester notes. The output is ready to paste into a pentest report. Example: cve_id='CVE-2024-3400', target='fw.corp.example.com', notes='Confirmed RCE via GlobalProtect gateway'.

Search recent FDA Office of Inspections and Investigations reading-room records by company, FEI, record type, country, establishment type, or publish date. This is official FDA document-index metadata with direct links to the posted records, plus incremental extracted document text when available, useful for finding recent 483-style inspection evidence by account.

Gets company profile information explaining who Walnai is, including legal structure, ownership, management, team background, and company experience.

Get full content of a specific BoardGameGeek forum thread, including all posts and replies. Use this after finding relevant threads with bgg-rules.

Remove the background from an image, leaving only the subject with transparency. Cost: $0.10. Requires Tier 1+.

Search for news articles using Bing News. Returns news results with titles, sources, thumbnails, and publication dates. Good for finding recent news coverage and trending stories.

Search prediction markets on Polymarket by topic. Returns matching markets with probabilities, volume, liquidity, and order book metrics. Good for finding specific events (elections, crypto prices, sports, geopolitics).

List all MCP servers by a single creator, plus aggregate trust signals. Use to evaluate a publisher holistically: 'do they ship consistently?', 'what's their security track record?', 'are there other servers by the same author?'. Match is case-insensitive on display name. Returns aggregate stats (total servers, avg security score, grade distribution, critical-finding count) plus the per-server list.

Find knowledge base entries similar to a given entry by comparing tags and content. Returns related contexts ranked by similarity score. Useful for discovering related patterns, examples, or documentation after finding one relevant entry.

Search 2,583 bookable Hawaii tours and activities by keyword, island, price range. Returns tours from Viator, GetYourGuide, Klook, and Groupon with affiliate booking links. Use this when users ask about Hawaii tours, activities, or things to do.

Import a job from a URL (e.g., LinkedIn, Greenhouse, Lever, Workday) and add it to your applications. Optionally trigger auto-apply immediately. Use this when a user has a direct link to a job posting.

List all Secureship API endpoints, optionally filtered by category. Categories include: carriers, shipment, history, address-book, auth, account, users, invoices, pickups, documents, integrations, settings, dashboard. Use for discovery when you don't know which endpoint to search for.

Get educational attainment for counties (population 25+). Returns counts and percentages for: less than high school, high school diploma/GED, some college/associate degree, bachelor's degree, and graduate/professional degree. Args: state: Two-letter state abbreviation (e.g. 'WA', 'CA') or 2-digit FIPS code. county_fips: Three-digit county FIPS code (e.g. '033' for King County). Omit to get all counties in the state. year: ACS 5-year estimate year (default 2022).

Analyze non-email messages (SMS, WhatsApp, Instagram DMs, Discord, Slack, Telegram, LinkedIn, Facebook Messenger, iMessage, Signal) for platform-specific threats including smishing, wrong-number scams, OTP interception, impersonation, and crypto fraud. $0.01/call via skyfire-api-key header (Skyfire Buyer API Key). By using this service you accept the Terms of Service. Advisory service only.

List HODLer Airdrop campaigns with optional filters (public; logged-in users may see extra participation info)

Search for land parcels by parcel ID prefix (autocomplete). Returns matching parcels with their district, area, and GPS coordinates. Useful for finding exact parcel IDs, then searching transactions nearby. Example: search for parcels starting with '146518_8.01'.

Check social media handle availability on GitHub, Twitter/X, TikTok, LinkedIn, and Instagram.

Remove the background from an image. Uses BiRefNet segmentation to precisely separate foreground from background. Returns a base64-encoded image with transparent background (PNG) or white background (WebP). Sub-500ms latency on GPU. Args: image_base64: Base64-encoded image data (PNG, JPEG, or WebP). output_format: Output format -- 'png' (with transparency) or 'webp'. Returns: dict with keys: - image_base64 (str): Base64-encoded result image - format (str): Output image format - original_size (dict): Original width and height - processing_ms (int): Processing time in milliseconds

Get recruiters who posted or are associated with a specific job. Returns contact info including email and LinkedIn.

Explains a specific security finding by its rule ID. Returns why it is a risk, how to fix it, and false positive considerations.

Delete existing role by id - You can't delete a role if it is still assigned to a user. In such cases please remove the role from the users.

List available background music tracks with names and durations. Use the exact musicName when creating shorts.

Remove image background with AI. Opens MiOffice in your browser to process the file privately — files never leave your device.

Remove the background from an image. Downloads the image from the provided URL, removes its background using AI, and returns the processed image. Args: image_url: URL of the image to process. Must be a publicly accessible image URL (PNG, JPEG, or WebP). Returns: The processed image with background removed, plus a summary message.