Official description7 sourcesMCP servers8d ago
afshinator/mcp-server-pexels
Search Pexels stock imag
View on GitHub ↗Direct link · no redirect page
Config template (adjust for your client)
claude mcp add mcp-server-pexels -- npx -y mcp-server-pexels
A pasteable skeleton generated from the repository name. Check the project README for the exact package and arguments.
- Star count
- 2 ★ · afshinator/mcp-server-pexels · read from the GitHub search API · (checked 2026-10-16 20:00:20)
- Source
- collected from the punkpeye/awesome-mcp-servers list · type: MCP server
- Harvest history
- first seen 2026-10-09 · last seen 2026-10-16 · seen 32× · listed by 7 sources
- Score
- score 703 · 2 ★ · 7 sources · mcp
FAQ (answers come from our records)
How do I install it?
We did not find an install command for this repository (our install coverage is only about 4%). Check its README or releases page.
Who is it for?
We filed it under these tags: MCP 服务. The tags come from the entry itself, not a taxonomy we imposed.
Who maintains it?
repository
afshinator/mcp-server-pexels；first catalogued 2026-10-09；last seen 2026-10-16。There is no official certification or quality rating here — we record sources and dates only.Third-party quality signals
mcp-server-pexels
以下摘录自该仓库的 README（片段） · 在 GitHub 读完整 README →
An MCP server for Pexels stock photo and video search. Optimized for LLMs.
Pexels provides free stock photos and videos.
Features
- **Photo Search** — Search photos with filters (query, orientation, size, color, locale)
- **Video Search** — Search videos, auto-selects HD .mp4 closest to 1920x1080
- **Get Details** — Retrieve full metadata for a photo/video by ID
- **Intelligent Caching** — 10 min TTL for searches, 60 min for ID lookups
- **Error Handling** — Graceful failures with helpful messages (per MCP best practices)
- **Attribution** — Mandatory photographer credits in every result
> **For LLM agents:** See docs/agents/AGENTS.md for agent-optimized documentation — tool semantics, caching behaviour, response structure, and attribution requirements.
Prerequisites
- Node.js v20+
- Pexels API key (free at pexels.com/api)
Quick Start (2 minutes)
1. Get an API Key
Sign up at pexels.com/api — free, no credit card.
2. Build the Server
3. Add to Claude Desktop
Open ~/Library/Application Support/Claude/claude_desktop_config.json (Mac) or %APPDATA%\Claude\claude_desktop_config.json (Windows), add:
> **Windows note:** Use node.exe full path or add Node to PATH. Forward slashes in paths work on Windows.
4. Restart Claude Desktop
The server is now available as pexels_search_photos, pexels_search_videos, and pexels_get_details.
Configuration
Add to your .mcp.json or claude_desktop_config.json:
Environment
Set PEXELS_API_KEY in your environment. For local dev, create a .env file:
Development
LLM-optimized docs for agent consumers are at docs/agents/AGENTS.md.
Tools
| Tool | Description |
|------|-------------|
| pexels_search_photos | Search for photos by query |
| pexels_search_videos | Search for videos |
| pexels_get_details | Get details by ID and type |
Architecture
- src/index.ts — Entry point, MCP server setup
- src/tools/ — Tool implementations
- src/shared/ — Cache, API client, errors, types, video selector
- src/utils/ — Zod validation schemas
Engineering Decisions
| Decision | Rationale |
|----------|----------|
| **Cache-first architecture** | Pexels API allows 200 requests/hour. Caching (10m TTL for searches, 60m for ID lookups) preserves quota, reduces latency to <5ms on cache hit, and demonstrates awareness of API costs — critical for production AI systems where agents frequently re-request the same context. |
| **Fail-fast at call time** | MCP servers are spawned as child processes — starting is not the time to fail. Server warns on startup but fails gracefully on first tool call with structured isError: true. |
| **Zod validation schemas** | MCP v2 SDK requires z.object() wrappers. Catches invalid input before it reaches the API. |
| **resource_link for media** | Remote images and videos are provided as MCP resource_link content blocks with proper mimeType. The markdown image link in the text block remains as a fallback for clients that do not render resource_link. |
Pexels provides free stock photos and videos.
Features
- **Photo Search** — Search photos with filters (query, orientation, size, color, locale)
- **Video Search** — Search videos, auto-selects HD .mp4 closest to 1920x1080
- **Get Details** — Retrieve full metadata for a photo/video by ID
- **Intelligent Caching** — 10 min TTL for searches, 60 min for ID lookups
- **Error Handling** — Graceful failures with helpful messages (per MCP best practices)
- **Attribution** — Mandatory photographer credits in every result
> **For LLM agents:** See docs/agents/AGENTS.md for agent-optimized documentation — tool semantics, caching behaviour, response structure, and attribution requirements.
Prerequisites
- Node.js v20+
- Pexels API key (free at pexels.com/api)
Quick Start (2 minutes)
1. Get an API Key
Sign up at pexels.com/api — free, no credit card.
2. Build the Server
3. Add to Claude Desktop
Open ~/Library/Application Support/Claude/claude_desktop_config.json (Mac) or %APPDATA%\Claude\claude_desktop_config.json (Windows), add:
> **Windows note:** Use node.exe full path or add Node to PATH. Forward slashes in paths work on Windows.
4. Restart Claude Desktop
The server is now available as pexels_search_photos, pexels_search_videos, and pexels_get_details.
Configuration
Add to your .mcp.json or claude_desktop_config.json:
Environment
Set PEXELS_API_KEY in your environment. For local dev, create a .env file:
Development
LLM-optimized docs for agent consumers are at docs/agents/AGENTS.md.
Tools
| Tool | Description |
|------|-------------|
| pexels_search_photos | Search for photos by query |
| pexels_search_videos | Search for videos |
| pexels_get_details | Get details by ID and type |
Architecture
- src/index.ts — Entry point, MCP server setup
- src/tools/ — Tool implementations
- src/shared/ — Cache, API client, errors, types, video selector
- src/utils/ — Zod validation schemas
Engineering Decisions
| Decision | Rationale |
|----------|----------|
| **Cache-first architecture** | Pexels API allows 200 requests/hour. Caching (10m TTL for searches, 60m for ID lookups) preserves quota, reduces latency to <5ms on cache hit, and demonstrates awareness of API costs — critical for production AI systems where agents frequently re-request the same context. |
| **Fail-fast at call time** | MCP servers are spawned as child processes — starting is not the time to fail. Server warns on startup but fails gracefully on first tool call with structured isError: true. |
| **Zod validation schemas** | MCP v2 SDK requires z.object() wrappers. Catches invalid input before it reaches the API. |
| **resource_link for media** | Remote images and videos are provided as MCP resource_link content blocks with proper mimeType. The markdown image link in the text block remains as a fallback for clients that do not render resource_link. |
- Source
- awesome · https://github.com/afshinator/mcp-server-pexels
- Browse by field
- MCP servers
- Traction
- 2 ★
- added
- yesterday
More in MCP servers
Filed under
Every tag opens the full list for that filter. Counts are the live catalogue figures.