snapmcp
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| SNAPMCP_DIR | No | /app/captures | |
| SNAPMCP_BADGE | No | false | |
| SNAPMCP_THEME | No | auto-detected | |
| SNAPMCP_FORMAT | No | png | |
| SNAPMCP_SHADOW | No | none | |
| SNAPMCP_PADDING | No | 32 | |
| SNAPMCP_QUALITY | No | 90 | |
| SNAPMCP_ALLOWED_PATHS | No | (deny-all) | |
| SNAPMCP_BORDER_RADIUS | No | 0 | |
| SNAPMCP_WINDOW_CHROME | No | false | |
| SNAPMCP_CHROME_CHANNEL | No | stable | |
| SNAPMCP_CHROME_PROFILE | No | ||
| SNAPMCP_SSRF_PROTECTION | No | true | |
| SNAPMCP_CHROME_EXECUTABLE | No |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
Use snapmcp to turn content into visual documentation artifacts (PNG/JPEG, PDF, GIF, Markdown/HTML/PPTX/DOCX). Select one focused tool per artifact; use capture_batch only for independent captures. Every tool returns a structured payload with the saved path, format and size. Use snapmcp-hint with a natural-language query or read snapmcp://capabilities when the correct tool is unclear. URL captures are SSRF-protected; file reads require an explicit allowlist.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": true
} |
| resources | {
"listChanged": true
} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| capture_terminalA | Generate a styled terminal screenshot from text lines. Uses a detected terminal palette when available (Kitty, GNOME, Alacritty, WezTerm, and others) and otherwise falls back to the configured theme. Use for CLI tutorials, command output documentation, and terminal-based guides. |
| capture_codeA | Generate a syntax-highlighted code screenshot using Shiki (300+ bundled languages and 27 themes). Supports inclusive startLine/endLine ranges and optional window chrome. Ideal for code documentation, tutorials, and sharing snippets with an IDE-like appearance. |
| capture_browserA | Take a screenshot of an HTTP(S) URL using headless Chromium. Uses Playwright's bundled Chromium by default; set SNAPMCP_CHROME_EXECUTABLE or the experimental SNAPMCP_CHROME_PROFILE to opt into a system Chrome or persistent profile. Explicit profile mode can send profile cookies and session state to the target site. SSRF protection is enabled by default and blocks private/local targets. Supports full-page or viewport captures. |
| capture_fileA | Read a file and generate a syntax-highlighted screenshot with automatic language detection from file extension. Supports 300+ bundled Shiki languages. Requires SNAPMCP_ALLOWED_PATHS to be set for security (deny-all by default). Maximum file size 5MB. |
| capture_markdownA | Render GitHub-flavored Markdown as a styled document screenshot. Supports tables, task lists, and fenced code blocks with syntax highlighting. Raw HTML is escaped, JavaScript is disabled, and external resources are blocked. Maximum input 200KB. |
| capture_htmlA | Render HTML/CSS as a screenshot in an isolated, JavaScript-disabled context. Inline styles and embedded data images are supported; scripts, frames, external stylesheets, and other external resources are removed or blocked. Maximum input 200KB. |
| capture_diffA | Render a unified diff with color-coded additions (green) and deletions (red). Shows file headers, hunk markers, and context lines. Ideal for PR reviews, migration guides, and change documentation. Maximum input 500KB. |
| capture_pdfA | Convert a URL to a print-quality PDF using headless Chromium. The complete paginated document is printed. Uses a system Chrome profile only when explicitly configured. SSRF protection is enabled by default (blocks private IPs and localhost). |
| capture_batchA | Capture multiple items in a single call. Each capture is processed sequentially with its own parameters. Use for batch documentation generation (e.g., capture terminal output, code file, and browser screenshot together). Maximum 10 captures per call. |
| capture_sequenceA | Capture ordered process steps as individual image files and optionally compile them into a GIF. Each step has typed capture parameters plus stepNumber and label. When GIF output is requested, frames are rendered as PNG. Maximum 60 steps. |
| capture_gifA | Create an animated GIF from 2-60 sequential captures. Frames are rendered as PNG, compiled into a GIF, and removed after encoding. loop=true repeats forever; loop=false plays once. Maximum canvas 8192x8192. |
| capture_documentA | Create a document with embedded step-by-step captures. Each capture is rendered as an image and embedded in the document with optional captions. Output formats: markdown (with image references), html (self-contained with base64 images), pdf (print-quality), pptx (editable PowerPoint deck), or docx (editable Word document). Maximum 30 captures per document; 25 MB per embedded image and 75 MB total. |
| snapmcp-hintA | Find the right snapmcp tool by natural-language intent, or get concise setup, security, sandbox, agent, and token guidance. Use when tool selection is unclear or before loading the full documentation. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
| snapmcp-capabilities | Compact, machine-readable catalog of snapmcp tools, limits, discovery surfaces, security defaults, and token policy. |
TDQS
Scored across 13 tools
Most tools are cleanly separated by input type (terminal, code, file, browser, markdown, html, diff, pdf). The trio capture_sequence / capture_gif / capture_batch overlaps conceptually (all orchestrate multiple captures), and capture_code vs capture_file both produce highlighted code images, though descriptions clarify input source and output form.
Twelve of thirteen tools follow a clear capture_<target> snake_case pattern that is easy to scan and predict. The single outlier, snapmcp-hint, uses a hyphenated server-prefixed name and a non-verb style, breaking the otherwise uniform convention.
13 tools is well within a healthy range for a multi-format screenshot/rendering server. Each tool maps to a distinct render target or orchestration mode, and none feels redundant enough to remove.
The surface covers essentially every common capture source: terminal, code, file, browser, markdown, html, diff, pdf, plus GIF, sequence, document, and batch composition. Remaining gaps are minor, e.g. no plain-text capture, no image post-processing/annotation, and no output listing or cleanup tool.