mcp-freesound
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| FREESOUND_API_KEY | Yes | Your Freesound API key (Client secret/Api key). Required. | |
| MCP_FREESOUND_HOST | No | HTTP host to bind when using the HTTP transport. | 127.0.0.1 |
| MCP_FREESOUND_PATH | No | HTTP path for the MCP endpoint when using the HTTP transport. | /mcp |
| MCP_FREESOUND_PORT | No | HTTP port to bind when using the HTTP transport. | 8000 |
| MCP_FREESOUND_CONFIG | No | Path to the configuration file. Defaults to the OS-specific config location. | per-OS config path |
| MCP_FREESOUND_TIMEOUT | No | Request timeout in seconds. | 30 |
| MCP_FREESOUND_TLS_KEY | No | Path to a TLS private key file for enabling HTTPS. | |
| MCP_FREESOUND_TLS_CERT | No | Path to a TLS certificate file for enabling HTTPS. | |
| MCP_FREESOUND_LOG_LEVEL | No | Log level for the server. | INFO |
| MCP_FREESOUND_TRANSPORT | No | Transport to use: stdio or http. | stdio |
| MCP_FREESOUND_AUTH_TOKEN | No | Bearer token required when using the HTTP transport. Generate one with `mcp-freesound token`. | |
| MCP_FREESOUND_DOWNLOAD_DIR | No | Folder where preview downloads are saved. | ~/Downloads/freesound |
| MCP_FREESOUND_ALLOWED_ORIGINS | No | Extra allowed origins for HTTP Origin checks. Defaults to localhost only. | localhost only |
| MCP_FREESOUND_ALLOW_DOWNLOADS | No | Whether the download_preview tool is allowed. | true |
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": false
} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| search_soundsA | Search Freesound for audio files (sound effects, ambiences, music loops...) by text. Supports Freesound filter syntax (e.g. 'duration:[0 TO 30] license:"Creative Commons 0"' or 'tag:rain type:wav') and sorting. Returns metadata, license and preview URLs. |
| get_soundB | Get full details of a single Freesound sound: technical properties (sample rate, bit depth, channels, size), license, pack, geotag, ratings, downloads, previews and waveform images. |
| find_similar_soundsC | Find sounds that are acoustically similar to a given Freesound sound. |
| get_sound_analysisA | Get audio descriptors computed by Freesound for a sound (e.g. bpm, key, loudness, brightness, pitch). Optionally restrict to specific descriptor names. |
| get_sound_commentsA | Get user comments posted on a Freesound sound (newest first). |
| get_userB | Get the public profile of a Freesound user (number of sounds, packs, about text, home page). |
| get_user_soundsB | List the sounds uploaded by a Freesound user. |
| get_user_packsB | List the packs (collections of sounds) created by a Freesound user. |
| get_packB | Get details of a Freesound pack (name, description, author, number of sounds). |
| get_pack_soundsC | List the sounds contained in a Freesound pack. |
| get_attributionA | Build ready-to-paste attribution/credits text for one or more Freesound sounds, with license details (whether attribution is required and whether commercial use is allowed). |
| download_previewA | Download the preview audio file (MP3 or OGG) of a Freesound sound to the configured download folder on the server machine. Full-quality originals require OAuth2 and are not supported; previews are free to fetch. |
| get_api_usageA | Show how many Freesound API requests the configured key has made recently, useful to monitor rate limits. |
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 13 tools
Each tool targets a distinct resource+action pair: sound detail, sound analysis, sound comments, acoustic similarity, user profile, user sounds, user packs, pack detail, pack contents. The only surface-level lookalikes are search_sounds vs find_similar_sounds, but the acoustic-similarity purpose is explicitly stated and clearly differentiated.
All names follow a predictable verb_noun snake_case pattern with consistent resource-noun nesting (get_sound_* / get_user_* / get_pack_* / *_sounds). The few alternative verbs (search_sounds, find_similar_sounds, download_preview) map cleanly to their distinct actions rather than breaking the convention.
13 tools sit comfortably in the well-scoped 3-15 range and each one covers a distinct facet of the Freesound API (search, metadata, analysis, people, packs, licensing, download, quota). Nothing appears redundant or filler.
The read-oriented domain is well covered end to end: discovery, detail, analysis, comments, similarity, users, packs, attribution, preview download, and rate-limit monitoring. Gaps are minor and mostly out-of-scope by design (no comment posting, no bookmarks, no full-quality download due to OAuth2).