Infomaniak AI
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| INFOMANIAK_API_TOKEN | No | Infomaniak API token used to start the server. Required unless INFOMANIAK_API_TOKEN_FILE is provided. | |
| INFOMANIAK_MCP_SHA256 | No | SHA256 checksum of the reviewed backend entry file for the original three-tool read-only edition. | |
| INFOMANIAK_API_TOKEN_FILE | No | Absolute path to a regular secret file (mode 0600) containing the Infomaniak API token. Do not set both token sources. | |
| INFOMANIAK_MCP_SERVER_PATH | No | Absolute path to the reviewed dist/server.js backend for the original three-tool read-only edition. |
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 | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| infomaniak_statusC | Check catalogue coverage, connection policy and reviewed capabilities. |
| infomaniak_searchB | Discover public API operations. Catalogue entries may be unavailable; reviewed schemas load on demand. |
| infomaniak_catalogueB | Browse all public API metadata by product/method. Discovery does not grant execution. |
| infomaniak_readC | Call a reviewed read within this connection's account/resource grants. |
| infomaniak_docs_searchA | Search current official Infomaniak help guides for Manager/browser tasks, or the live developer API index. Use public feature terms only: never include customer identifiers, addresses, credentials or private context. Results are documentation, not permission or execution coverage. |
| infomaniak_docs_readA | Fetch a current official help article or developer API reference. Returns source, retrieval time, steps/links and explicit pagination; no provider token or browser cookies. Use the guide with your own authorized browser/computer tools when an API capability is missing. Never bypass denied scopes, pending approvals or resource locks. |
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 6 tools
The docs_search/docs_read pair is clearly distinguished by action, but infomaniak_catalogue and infomaniak_search largely overlap — both describe discovering/browsing public API metadata and operations, and their descriptions even cross-reference the same catalogue. An agent could plausibly pick either for API discovery, though the docs-vs-live capability split is somewhat explained in the text.
All names use the same infomaniak_ prefix and snake_case, with a recognizable docs_search/docs_read and search/read pairing. Minor deviation: some suffixes are verbs (search, read) while others are nouns (catalogue, status), which slightly breaks the verb_noun expectation.
Six tools is well-scoped for a documentation-discovery plus read-execution server, and each maps to a plausible step (search, browse, read docs, execute read, check status). Slightly thin — one redundant discovery tool means the effective distinct surface is closer to five.
The surface covers discovery, documentation retrieval, capability introspection, and read execution, but has no write/create/update/delete operations at all, so any mutation workflow dead-ends. The read-only scope appears intentional and is documented as a policy boundary, which mitigates but does not eliminate the gap for an API-access server.