RenderProof
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| JINA_API_KEY | No | API key for Jina reader. | |
| RENDERPROOF_MAX_CHARS | No | Maximum characters for text output. Default: 12000 | |
| RENDERPROOF_OUTPUT_DIR | No | Directory for output evidence files. Default: .renderproof/evidence | |
| RENDERPROOF_ALLOWED_HOSTS | No | Comma-separated list of allowed hosts. Default: (none, public URLs only) | |
| RENDERPROOF_ALLOW_PRIVATE_NETWORK | No | Set to '1' to allow private network access. | |
| RENDERPROOF_ENABLE_REMOTE_READERS | No | Set to '1' to enable remote readers. |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": true
} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| route_web_taskA | Choose the cheapest reliable web perception route before reading, browsing, or using screenshots. |
| read_urlA | Read a public http(s) URL as normalized text, using direct fetch first and optional remote readers only when configured. |
| capture_pageB | Capture rendered page evidence with Playwright when visual layout, screenshots, or browser state matter. |
| capture_motionA | Record short rendered page motion evidence with Playwright video when animations, transitions, scrolling, or dynamic media matter. |
| analyze_motionC | Analyze rendered page motion for coding agents using sampled keyframes, CSS animation metadata, pixel diffs, and a contact sheet. |
| clone_websiteA | Generate a clone-ready evidence bundle for a public website: desktop/mobile screenshots, design tokens, assets, page topology, behaviors, and component specs. Use only for legitimate site migration, learning, or authorized rebuilds. |
| doctorB | Check local runtime, security policy, optional remote readers, and Playwright availability. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 7 tools
Each tool has a distinct role: routing decisions, text reading, static page capture, motion capture, motion analysis, runtime diagnostics, and full site cloning. There is no overlap between tools; capture_page and capture_motion are clearly differentiated by media type, and analyze_motion builds on captured motion data rather than duplicating it.
Most tools follow a clear verb_noun pattern (read_url, capture_page, capture_motion, analyze_motion, clone_website). The 'doctor' tool is a slight outlier as a noun-based name, but it's memorable and contextually acceptable. Overall the naming convention is predictable and consistent enough.
Seven tools is well within the ideal range for a focused web perception and cloning utility. Each tool adds a distinct capability without bloat, and the count matches the server's stated purpose of routing, reading, capturing, analyzing, and cloning web content.
The toolset covers the full lifecycle of web task handling: route_web_task plans the approach, read_url extracts content, capture_page and capture_motion provide visual evidence, analyze_motion gives deeper insights, doctor ensures environment health, and clone_website produces a complete bundle. There are no obvious missing core operations for this domain.