Which integrations are available for this server?

Generates Instagram post thumbnail images from a title, with configurable format, model (SD/HD), output format (WebP/PNG), and content category. Generates YouTube thumbnail images from a title, with configurable format, model (SD/HD), output format (WebP/PNG), and content category.

mcp-thumbapi

by dinalllll

Overview Schema Related Servers Score Discussions

JavaScript

Local

@thumbapi/mcp-server

MCP server that exposes the ThumbAPI thumbnail generation endpoint as a Model Context Protocol tool. Point any MCP-compatible AI agent at it and ask for a YouTube thumbnail, Instagram post, X/Twitter card, LinkedIn share, or blog hero image from a title.

Transport: stdio (local, no remote server)
Runtime: Node.js 18+, installed via npx
Tools exposed: generate_thumbnail, login, logout

The first time you use the MCP, ask your client (Claude Desktop, Cursor, Claude Code, Windsurf, Cline, Continue) to "log in to thumbapi" — that triggers the login tool. It:

Spins up a one-shot local callback server on a random loopback port.
Opens your browser to app.thumbapi.dev/mcp-login.
Waits for you to click Authorize on the page (log in first if needed).
Writes your API key to ~/.thumbapi/config.json (mode 0600, dir 0700).

Every subsequent generate_thumbnail call reads from that file. No copy-paste.

To sign out locally, ask your client to "log out of thumbapi" — that runs the logout tool, which deletes ~/.thumbapi/config.json. The key on the ThumbAPI dashboard is untouched (rotate it in the dashboard if you also want to invalidate the key server-side).

Related MCP server: RendrKit

Install & configure

Below are configs for the 5 most common MCP clients. Each uses the same command and args — only the config file location and JSON wrapper shape change per client. No API key goes into the config; the login tool handles that on first use.

Claude Desktop

Config file:

macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
Windows: %APPDATA%\Claude\claude_desktop_config.json

{
  "mcpServers": {
    "thumbapi": {
      "command": "npx",
      "args": ["-y", "@thumbapi/mcp-server"]
    }
  }
}

Restart Claude Desktop after saving.

Cursor

Config file: ~/.cursor/mcp.json (or per-project .cursor/mcp.json)

{
  "mcpServers": {
    "thumbapi": {
      "command": "npx",
      "args": ["-y", "@thumbapi/mcp-server"]
    }
  }
}

Reload Cursor's MCP servers from Settings → MCP.

Windsurf

Config file: ~/.codeium/windsurf/mcp_config.json

{
  "mcpServers": {
    "thumbapi": {
      "command": "npx",
      "args": ["-y", "@thumbapi/mcp-server"]
    }
  }
}

Open Windsurf → Settings → Cascade → MCP Servers → Refresh.

Cline (VS Code extension)

Config file:

macOS: ~/Library/Application Support/Code/User/globalStorage/saoudrizwan.claude-dev/settings/cline_mcp_settings.json
Windows: %APPDATA%\Code\User\globalStorage\saoudrizwan.claude-dev\settings\cline_mcp_settings.json
Linux: ~/.config/Code/User/globalStorage/saoudrizwan.claude-dev/settings/cline_mcp_settings.json

{
  "mcpServers": {
    "thumbapi": {
      "command": "npx",
      "args": ["-y", "@thumbapi/mcp-server"]
    }
  }
}

Open the Cline sidebar → MCP Servers → Restart.

Continue.dev

Config file: ~/.continue/config.json

{
  "experimental": {
    "modelContextProtocolServers": [
      {
        "transport": {
          "type": "stdio",
          "command": "npx",
          "args": ["-y", "@thumbapi/mcp-server"]
        }
      }
    ]
  }
}

Reload Continue from the VS Code / JetBrains extension.

Tool reference

`generate_thumbnail`

Generates a thumbnail from a title.

Param	Type	Required	Notes
`title`	string (1–200 chars)	yes	The headline / video title.
`format`	`youtube` \| `instagram` \| `x` \| `blogpost` \| `linkedin`	yes	Target platform / aspect ratio.
`model`	`sd` \| `hd`	no	Default `sd` (10 credits). `hd` needs Pro/Business (20 credits).
`outputFormat`	`webp` \| `png`	no	Default `webp`.
`category`	one of the values below	no	Content category hint that biases visual style. Omit to auto-detect from the title.

category accepted values — the schema constrains this to the list below; the LLM should pick the closest match, or leave category unset:

auto, tech-saas, business-finance, education-tutorial, fitness-wellness, medical-healthcare, lifestyle-vlog, food-cooking, travel, gaming, entertainment-comedy, news-commentary, creative-design.

Returns an MCP image content block plus a text summary that includes:

generationId — stable ID for the generation (useful for logs and audits).
imageUrl — public URL on ThumbAPI's CDN. Use this to download or embed the image without decoding base64. The URL is served from Cloudflare R2 and is returned on every successful generation (v1.1.0+).

`login`

Signs the MCP server in to ThumbAPI. Starts a local callback server on a random loopback port, opens your browser to https://app.thumbapi.dev/mcp-login?callback=…&state=…, and waits for you to click Authorize on the page. The returned API key is written to ~/.thumbapi/config.json (mode 0600, directory 0700).

Takes no arguments.
Blocks up to ~30 seconds per invocation waiting for the browser callback. If you haven't consented yet, the tool returns "Waiting for browser approval…" — just call login again to keep polling. The underlying local server stays alive across calls for up to 15 minutes.
Idempotent: calling login again while a flow is in progress reuses the same URL / port.
Only accepts callback traffic bound to 127.0.0.1 and requires the CSRF state param to match — no other machine on your LAN can hit the callback.

`logout`

Deletes ~/.thumbapi/config.json. Takes no arguments. Does not revoke the key server-side — rotate it in the dashboard if you want to invalidate the key everywhere.

License

MIT

Available Tools

3 tools

generate_thumbnailA

Generate a thumbnail / social share image from a title using the ThumbAPI service.

Use this when the user wants to create a YouTube thumbnail, Instagram post image, X/Twitter card, LinkedIn share image, or blog post hero image from a headline or title.

Returns the generated image inline (viewable by the model) plus metadata: format, outputFormat, generationId, and imageUrl (a public URL on ThumbAPI's CDN — use this to download or embed the image without handling base64).

Requires an API key. Either set THUMBAPI_API_KEY in the environment or run the login tool once — that starts an OAuth-style browser flow and saves the key to ~/.thumbapi/config.json. Get a key at https://thumbapi.dev.

ParametersJSON Schema

Name	Required	Description
`model`	No	Generation model. `sd` (default) is faster and cheaper (10 credits). `hd` produces higher quality output but requires a Pro or Business plan (20 credits).
`title`	Yes	The headline / video title / blog post title to render on the thumbnail. Required. Max 200 chars. Write the title exactly as it should appear.
`format`	Yes	Target platform / aspect ratio. `youtube` = 1280x720 (16:9), `instagram` = 1080x1080 (1:1), `x` = 1600x900 (16:9, Twitter/X card), `blogpost` = photorealistic hero image (16:9), `linkedin` = 1200x627 (LinkedIn share).
`category`	No	Optional content category hint that biases visual style. Choose the closest match, or omit to let the API auto-detect from the title (equivalent to `auto`).
`outputFormat`	No	Image encoding of the returned file. Defaults to `webp`.

TDQS

A4.4/5.0

Behavior5/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

With no annotations, the description carries the full burden and does an excellent job: it discloses the returned data (inline image plus metadata fields), the public CDN URL, the authentication requirement, and the account-plan constraints for `hd` models. This is far beyond typical descriptions.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The description is well-structured with clear sections (purpose, usage, return, auth) and is not excessively long. It earns its length by covering both usage and behavioral details, though it could be slightly shorter by moving auth instructions elsewhere.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness5/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

Given the tool has no output schema, the description compensates by explicitly describing the return format and metadata. It also covers prerequisites, costs, and model selection, making it self-contained for an agent to use correctly.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters3/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

The input schema already covers all parameters with rich descriptions, so the baseline is 3. The description adds some operational context (e.g., using `imageUrl` rather than base64), but it does not substantially elaborate on parameter meaning beyond the schema.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description opens with a specific verb ('Generate') and resource ('thumbnail / social share image'), and further clarifies by listing concrete platforms (YouTube, Instagram, etc.). This clearly distinguishes it from the auth-focused sibling tools `login` and `logout`.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines4/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

It explicitly states when to use the tool ('Use this when...') with a list of use cases, and it mentions the prerequisite of an API key and the `login` alternative for auth. However, it does not explicitly describe when not to use it or mention alternative thumbnail-generation tools.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

loginA

Sign the MCP server in to ThumbAPI. Opens a browser to app.thumbapi.dev/mcp-login, asks the user to consent, and saves the returned API key to ~/.thumbapi/config.json.

Call this tool when:

The user asks to log in / sign in / authenticate to ThumbAPI.
Another tool (generate_thumbnail) returns "no API key found".

The tool blocks up to ~30s per invocation waiting for the browser callback. If the user hasn't consented yet, it returns "still waiting" — just call login again to keep waiting. The underlying local callback server stays alive across calls for up to 15 minutes.

ParametersJSON Schema

Name	Required	Description	Default
No parameters

TDQS

A4.8/5.0

Behavior5/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

With no annotations, the description fully discloses behavior: blocks up to 30s, returns 'still waiting' if not yet consented, callback server persists for 15 minutes across calls, and saves key to a config file. This gives the agent actionable expectations about timeouts and retry semantics.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness5/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The description is well-structured with three logical sections: purpose/action, when to call, and behavioral notes. Each sentence adds necessary information without fluff. The bulleted usage guidelines improve scannability.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness4/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

The description covers what the tool does, when to call, and key edge cases (blocking, 'still waiting', server lifetime). However, since there is no output schema, it would be helpful to specify what a successful response looks like, not just the pending case. This is a minor gap that keeps it from a perfect score.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters4/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

The tool takes zero parameters, so the schema already fully covers parameter semantics (100% coverage). Baseline for 0 params is 4, and the description doesn't need to add parameter info. It appropriately omits any parameter discussion.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description clearly states the tool's purpose: 'Sign the MCP server in to ThumbAPI' with a specific verb and resource. It also provides concrete details about the process (opens browser, asks consent, saves API key), which distinguishes it from siblings like logout and generate_thumbnail.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines5/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

Explicit 'Call this tool when' section lists two clear triggering conditions: user requests authentication, or generate_thumbnail returns 'no API key found'. This provides unambiguous guidance on when to use the tool and implicitly contrasts with logout for the opposite action.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

logoutA

Remove the saved ThumbAPI API key from ~/.thumbapi/config.json. Does not revoke the key on the server — that's a separate action from the dashboard.

ParametersJSON Schema

Name	Required	Description	Default
No parameters

TDQS

A4.9/5.0

Behavior5/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

With no annotations, the description carries the full burden of behavioral disclosure. It openly states that the tool modifies a specific local file and explicitly notes that it does NOT revoke the key on the server, giving the agent a clear understanding of the tool's exact effects and limitations.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness5/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The description is two sentences that front-load the primary action and immediately follow with a critical clarification. Every word contributes meaning, with no redundancy or filler.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness5/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

The tool is simple (no parameters, no output schema), and the description covers all necessary context: what it does, where it operates, and what it intentionally does not do. This is a complete and self-contained description.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters4/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

The tool has zero parameters, so parameter explanation is unnecessary. Per the rubric, a baseline of 4 is appropriate when there are no parameters and the description is clear about the tool's function.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

Description states a specific action ('Remove the saved ThumbAPI API key from ~/.thumbapi/config.json') and clearly distinguishes itself from server-side revocation with the explicit note 'Does not revoke the key on the server.' This makes the tool's purpose unmistakable and separates it from sibling tools like login and generate_thumbnail.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines5/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

The description explicitly clarifies when this tool is appropriate (local logout) and when it is not (server-side revocation, which is a separate dashboard action). This provides clear usage boundaries and prevents misuse.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections. Dates show when Glama detected each change.

3 tool updatesv1.1.1Jul 31, 2026
- First observedgenerate_thumbnail
- First observedlogin
- First observedlogout

Install Server

RayrunBuild, deploy, and operate MCP servers without becoming an infrastructure team.ray.runAd

TDQS

A4.7/5.0

Disambiguation5/5

Each tool has a distinct purpose: generate_thumbnail for the core task, login for authentication, and logout for de-authentication. There is no functional overlap, so an agent can easily select the right tool.

Naming Consistency4/5

All tool names are lowercase with underscores, but the pattern is mixed: generate_thumbnail follows verb_noun, while login and logout are simple verbs. This is a minor deviation and the names are still intuitive.

Tool Count5/5

With 3 tools, the server is well-scoped for its purpose: one core operation plus necessary authentication lifecycle. This is a reasonable size and each tool serves a clear function.

Completeness5/5

The tool surface covers the full workflow: authenticate (login), generate thumbnail (generate_thumbnail), and de-authenticate (logout). No obvious missing operations for the stated domain of ThumbAPI.

Maintenance

ActivitySlowing

ResponsivenessNo issues

Resources

Need Help?

Related Servers

Unclaimed servers have limited discoverability.

Looking for Admin?

If you are the server author, to access and configure the admin panel.

Related MCP Connectors

NativeAds AOAuth
ai.nativeads.mcp
On-brand ad creative generation: teach it your brand once, generate images and video forever.
AI & Machine Learning Marketing
1
designforyou A
org.swapp1990.designforyou
Generate logos, social posts, app screenshots, comic panels & visual-novel assets from prompts.
Design Tools Image & Video Processing
412
PixelDrive
pro.pixeldrive
Generate branded images at scale: design templates and render them to PNGs by prompt.
Image & Video Processing Design Tools AI & Machine Learning
2
CanvoraOAuth
ai.canvora
Turn any idea, URL, doc, or PDF into on-brand visuals: 100+ formats, native in 150+ languages
Design Tools Marketing Workplace & Productivity
2

Related MCP Servers

Image Generation MCP
Image & Video Processing Marketing Social Media
12-days-of-shipmas-2025
A
license
A
quality
C
maintenance
Generates blog and social media images using Google's Gemini AI with pre-configured platform presets for Ghost, Medium, Instagram, Twitter, LinkedIn, YouTube, and more.
Updated 8 months ago (2025-12-31 11:21 UTC)
2
19
MIT
RendrKit
Image & Video Processing Marketing Social Media
vbiff
A
license
A
quality
C
maintenance
Generate production-ready images from text prompts. AI-powered design API with 50+ templates for social posts, banners, OG images, and more.
Updated 6 months ago (2026-02-26 08:55 UTC)
8
13
MIT
Canvora MCP
canvora
A
license
Not graded
quality
B
maintenance
Enables AI agents to generate on-brand visuals from ideas, URLs, documents, or PDFs in over 100 formats and 150+ languages, with consistent brand kits.
Updated 24 days ago (2026-08-13 11:02 UTC)
10
MIT
Article Writer MCP
Content Management Systems Version Control
Cappybara12
F
license
Not graded
quality
B
maintenance
Enables automated content creation pipeline: topic research, article writing, image sourcing, and GitHub PR submission for blogs.
Updated a month ago (2026-07-28 21:23 UTC)
1
-

Latest Blog Posts

Who's Calling? MCP Hosts Are an Identity Blind Spot (And the Spec Knows It)
By Om-Shree-0709 on July 25, 2026.
mcp
Agent Identity
OAuth 2.1
Your AI Chatbot Just Exposed Your CEO's Salary to an Intern
By Om-Shree-0709 on July 2, 2026.
Agent Identity
MCP Security
OAuth Delegation
Why MCP Servers Need Execution Sandboxing (And Why Your Current Stack Isn't Enough)
By Om-Shree-0709 on June 30, 2026.
Agentic Ai
Prompt Injection
WebAssembly

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/dinalllll/thumbapi-mcp-server'

If you have feedback or need assistance with the MCP directory API, please join our Discord server

@thumbapi/mcp-server

Sign in

Install & configure

Claude Desktop

Cursor

Windsurf

Cline (VS Code extension)

Continue.dev

Tool reference

generate_thumbnail

login

logout

License

Available Tools

Tool Schema Changelog

Maintenance

Resources

Looking for Admin?

Related MCP Connectors

Related MCP Servers

Latest Blog Posts

MCP directory API

`generate_thumbnail`

`login`

`logout`