Skip to main content
Glama
deenesjakoruzh
starter-series

profilekit-mcp

by starter-series
OverviewSchemaRelated ServersScoreDiscussions
TypeScript
Remote

profilekit-mcp

MCP server for ProfileKit. Build GitHub profile SVG cards through conversation — from Claude Code, Codex CLI, ChatGPT Apps, or any other MCP-capable agent.

Currently implemented

  • 3 tools over stdio MCP: list_cards, list_themes, render.

  • Card types covering data (stats, pin, leetcode, …), blog layout (hero, section, timeline, …), animations (typing, snake, matrix, …), composition (stack), and utility (health) — exact set returned live by list_cards.

  • Built-in themes (tokyo_night, kanagawa, rose_pine, dracula, nord, …) — pass ?theme=<name> to any card; exact set returned live by list_themes.

  • Dynamic catalog sync from https://profilekit.vercel.app/api/catalog, cached per process; falls back to a bundled snapshot if the fetch fails.

  • Custom palettes via ?theme_url=<gist-raw-url> (supported on /stats and /stack as of ProfileKit v1).

  • Package identity and CLI binary are both profilekit-mcp; the package exports a typed runServer entrypoint.

Related MCP server: unbiased

Planned

  • compose_readme(sections) tool — return a full blog-layout README snippet in one call.

  • Palette suggestion tool backed by the caller's own vision/LLM capability — no built-in model calls.

Design intent

  • URL-only, never inlines SVG. render returns a URL plus markdown / HTML snippets; the SVG is fetched by the eventual <img> consumer (GitHub, dev.to, Notion, …). Tool responses stay small, side-effect-free, and embeddable anywhere external images are allowed.

  • One MCP server, three agents. After OpenAI and Anthropic co-announced MCP Apps in early 2026, a single stdio server covers Claude Code + Codex CLI + ChatGPT Apps natively — no per-platform adapter.

  • Live catalog over hardcoded list. Card definitions live in ProfileKit's /api/catalog, so when ProfileKit ships a new card the MCP server picks it up without a republish. The bundled fallback exists only so cold/offline starts still work.

  • No ranking, composable presentation. Mirrors ProfileKit's stance — every card is an independent SVG that the user composes, not a leaderboard.

Non-goals

  • Inlining card SVG into tool responses. The MCP server intentionally does not fetch card content. Agents that need to reason over the markup can fetch the URL themselves.

  • Built-in model calls. Future "suggest a palette" or "describe this card" features delegate to the calling agent's own LLM — this server never makes outbound LLM API calls.

  • Ranking, leaderboards, or rendering opinions in the tool surface.

Redacted

(none for this repo)

Install

After the unscoped npm package is published:

npm install -g profilekit-mcp

Before npm publication, run from this repository:

npm ci
npm run build
node dist/bin.js help

Register with your agent

Claude Code — add to .claude/settings.json in your repo:

{
  "mcpServers": {
    "profilekit": { "command": "profilekit-mcp" }
  }
}

Codex CLI — add to ~/.codex/config.toml:

[mcp_servers.profilekit]
command = "profilekit-mcp"

ChatGPT Apps — (Apps SDK MCP adapter; see the Apps SDK docs for wire-up)

Usage

Inside any registered agent, just ask:

> What ProfileKit cards exist?
> Render a tokyo_night stats card for heznpc.
> Give me a hero banner saying "heznpc" with subtitle "Building the ecosystem AI lives in", wave background, space-grotesk font.
> Build a kanagawa-themed pin card for heznpc/ProfileKit.

The agent will invoke list_cards / list_themes / render under the hood and hand you back a URL + markdown snippet ready to paste into your README.

Verify locally

npm ci
npm audit --audit-level=high
npm test
npm run build
npm run smoke:mcp
npm run pack:check

npm run smoke:mcp builds the package, starts dist/bin.js over stdio through the MCP SDK client, lists the three tools, renders a deterministic card URL, and verifies required-param errors. npm run pack:check runs npm pack --dry-run --json and verifies the exported types, server export, CLI binary, and required package files are present in the tarball.

Tools

Tool

Description

list_cards

Enumerate every card type returned by the live catalog, with descriptions and required params

list_themes

List the built-in themes

render

Build a card URL + markdown + HTML snippet for a given type and params

Example conversation

You: Render a pin card for heznpc/anvil using the rose_pine theme.

Agent: [calls render(type="pin", params={username: "heznpc", repo: "anvil", theme: "rose_pine"})]

URL:
https://profilekit.vercel.app/api/pin?username=heznpc&repo=anvil&theme=rose_pine

Markdown:
![pin](https://profilekit.vercel.app/api/pin?username=heznpc&repo=anvil&theme=rose_pine)

HTML:
<img src="https://profilekit.vercel.app/api/pin?username=heznpc&repo=anvil&theme=rose_pine" alt="pin" />

License

MIT © heznpc

Available Tools

3 tools
list_cardsA

List every ProfileKit card type (stats, hero, snake, ...) with a one-line description and the required params for each. Use this before calling render when the user asks what cards exist or which to use. Catalog is fetched live from ProfileKit on first call and cached per process.

ParametersJSON Schema
NameRequiredDescriptionDefault

No parameters

TDQS

A4.7/5.0
Behavior4/5

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

No annotations exist, so the description must convey behavior. It states the catalog is fetched live from ProfileKit on first call and cached per process, which is useful context. A slight deduction for not mentioning if listing is read-only (though implied).

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?

Three sentences, each earning its place: first explains what it does, second when to use it, third a behavioral note. No wasted words.

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 zero parameters and no output schema, the description covers purpose, usage, and behavior completely. No gaps.

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 input schema has no parameters, and schema description coverage is 100% (vacuously). The description adds meaning by describing what the output contains (one-line description, required params) beyond the empty 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 states the tool lists every ProfileKit card type with a one-line description and required params. It specifies the exact resource (card types) and action (list), clearly distinguishing it from siblings like 'render'.

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 says to use this tool before calling 'render' when the user asks what cards exist or which to use. This provides clear context for when to use it versus alternatives.

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

list_themesA

List built-in ProfileKit themes (dark, tokyo_night, kanagawa, rose_pine, ...). Any card accepts ?theme=<name>. For fully custom palettes use ?theme_url= pointing to a JSON gist.

ParametersJSON Schema
NameRequiredDescriptionDefault

No parameters

TDQS

A4.2/5.0
Behavior3/5

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

With no annotations, description carries full burden. Discloses it lists only built-in themes, and hints at card integration. No mention of whether it returns names only or full details, but adequate for a simple list tool.

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?

Two sentences, front-loaded with purpose and examples, no fluff. Every word earns its place.

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?

For a parameterless list tool with no output schema, description is sufficient. Explains purpose, examples, and integration with cards. Could mention if output is sorted or filtered, but not critical.

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?

Schema has no parameters (coverage 100%), and description adds value by explaining how the output relates to card usage, providing context beyond the empty 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?

Clearly states it lists built-in ProfileKit themes and provides specific examples (dark, tokyo_night, kanagawa, rose_pine). Distinguishes from sibling tools like list_cards and render.

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?

Explains how to use themes with cards via ?theme= parameter, and mentions alternative ?theme_url= for custom palettes. Could be more explicit about when to choose each option.

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

renderA

Build a ProfileKit card URL plus ready-to-paste markdown and HTML snippets for the given card type and params. Does NOT fetch the SVG itself — the URL is what consumers embed. Call list_cards first if unsure which type or what params the user's card accepts.

ParametersJSON Schema
NameRequiredDescriptionDefault
typeYesCard type (e.g. 'stats', 'hero', 'snake'). Must be one of the keys returned by list_cards.
paramsNoCard-specific parameters as key/value pairs (e.g. {username: 'heznpc', theme: 'tokyo_night'}). See list_cards output for common params per type. Values are stringified and URL-encoded.
altNoOptional alt text for the markdown image. Defaults to the card type.

TDQS

A4.4/5.0
Behavior4/5

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

No annotations are provided, so the description carries the full burden. It clearly states the tool does NOT fetch the SVG, which is a key behavioral trait. However, it does not mention side effects, permissions, or rate limits, which would be needed for full transparency. The explicit negation of SVG fetching earns a high score, but some gaps remain.

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 long, highly concise, and front-loads the core purpose. Every sentence provides essential information without redundancy.

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?

Given the tool has no output schema and moderate complexity (nested params), the description adequately explains the tool's output (URL and snippets) and workflow (call list_cards first). It could mention the output format in more detail, but the context signals (no output schema) mean the description carries this burden well.

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?

Schema description coverage is 100%, so baseline is 3. The description adds minimal extra meaning beyond the schema: it explains that values are stringified and URL-encoded, and suggests using list_cards output for common params. This adds some value but not enough to raise the score above baseline.

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 builds a ProfileKit card URL and markdown/HTML snippets, specifying the verb 'Build' and the resource 'ProfileKit card URL plus snippets'. It distinguishes from 'list_cards' by noting it does NOT fetch the SVG, which helps clarify the tool's scope.

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 advises to call 'list_cards' first if unsure about the card type or parameters, providing clear when-to-use and when-not-to-use guidance. It also mentions that the URL is for embedding, not fetching SVG, setting correct expectations.

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.

  1. 3 tool updatesv0.2.1
    • First observedlist_cards
    • First observedlist_themes
    • First observedrender
Install Server
RayrunBuild, deploy, and operate MCP servers without becoming an infrastructure team.ray.runAd

TDQS

A4.3/5.0
Disambiguation5/5

Each tool has a distinct purpose: list_cards discovers available card types, list_themes retrieves theme options, and render generates card URLs. There is no overlap between them.

Naming Consistency4/5

Tools follow a consistent 'verb_noun' pattern with 'list_' for listing and 'render' for generation. 'render' is a bare verb while others have prefix, but it's minor.

Tool Count5/5

With 3 tools covering listing, theming, and rendering, the scope is tight and focused. Each tool is essential and the count is appropriate for a card generation server.

Completeness3/5

The tools cover the core workflow of discovering cards and themes and generating URLs, but there is no tool for updating or deleting cards (if such operations exist), nor for fetching the rendered SVG directly.

Maintenance

ActivityActive
ResponsivenessUnresponsive

Resources

Unclaimed servers have limited discoverability.

Looking for Admin?

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

Related MCP Connectors

Related MCP Servers

Latest Blog Posts

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/starter-series/profilekit-mcp'

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