AgentSearch Web Render
agentsearch_renderRender JavaScript-heavy pages in headless Chromium to capture fully loaded content as markdown, text, HTML, links, or screenshot when static extraction returns empty or partial results.
Instructions
Render one absolute http(s) URL in headless Chromium with AgentSearch Web Render, so JavaScript runs before content is captured, and return the backend response as pretty-printed JSON text (a non-JSON body is wrapped as {raw: text}). The response includes the final url, title, status_code, meta (render_ms, chars, truncated, js, lang …), warnings, error, and the rendered page body in each requested format: markdown (default), text, html, and/or links; plus an optional base64 screenshot. Use this when the page is a JavaScript-heavy app (SPA, client-rendered docs or dashboards) where agentsearch_extract returns empty or partial content; for static pages prefer agentsearch_extract (faster), and to find pages from a query use agentsearch_web_search. Paid tool: needs a wallet key (AGENTSEARCH_WALLET_PRIVATE_KEY or POCKET_PRIVATE_KEY) and then pays per call in USDC on Base over x402 through the Pocket Network Agentic Portal, never more than the per-call cap or AGENTSEARCH_MAX_TOTAL_ATOMIC per session (default $1.00); the result includes the payment receipt and the portal envelope {portal, data}, whose data is untrusted third-party content. With no wallet key it makes no network call and explains how to add one. If AGENTSEARCH_BASE_URL is set, it calls your self-hosted AgentSearch API instead (no payment). Note: the free tier serves search only; this tool needs a wallet key or a self-hosted API. Read-only: it loads the page like a browser visit and does not click, submit, or change anything, but it uses the network, has a hard backend deadline of about 4.2 s, and can hit rate limits (HTTP 429). Target-site failures (DNS, TLS, timeouts, bot walls, robots.txt) come back as a normal result with a non-null error object {code, message, retryable} and empty or partial content; invalid input or a disallowed URL (HTTP 400, e.g. UPSTREAM_REJECTED), rate limits (429) and service faults (5xx) are MCP errors with the code and whether it is retryable. Optional fields are forwarded unchanged to the API, which applies its own defaults and limits.
Input Schema
| Name | Required | Description | Default |
|---|---|---|---|
| url | Yes | Absolute public http(s) URL to render (port 80/443). | |
| block | No | Resource types to block while rendering (default ["media","font"]). | |
| formats | No | Output formats to return (default ["markdown"]). Add "html" for the rendered DOM, "text" for plain text, "links" for anchors. | |
| wait_ms | No | Extra fixed wait in ms after load (0–1500, default 0). | |
| viewport | No | Browser viewport size. | |
| full_page | No | Capture the full page height, clipped to 4000 px (default false). | |
| max_chars | No | Cap on returned characters per format (1–100000, API default 20000). | |
| screenshot | No | Also capture a screenshot, returned as base64 (default false). | |
| wait_until | No | Navigation event to wait for before capture (default domcontentloaded). | |
| screenshot_format | No | Screenshot encoding (default png; PNG over 1 MB is re-encoded as JPEG). | |
| wait_for_selector | No | CSS selector to wait for (at most 1.5 s) before capture. |