Zum Inhalt springen

Verlauf von mcp-server im Namensraum io.github

io.github.infino-ai/mcp-server · Registerstatus: aktiv

Die Quelle liefert zu diesem Eintrag keinen eigenen Namen. In der Überschrift steht deshalb der hintere Teil des Bezeichners; vollständig steht er als Koordinate darunter.

Angaben zu diesem Eintrag melden

Jede veröffentlichte Änderung an diesem Eintrag, die jüngste zuerst.

Änderungen
2026-10-02Deklarierte Version: 0.15.0 → 0.16.0
2026-10-02Im Register geändert am (Rohangabe): 2026-09-22 → 2026-10-01
2026-10-02Im Register eingetragen am (Rohangabe): 2026-09-22 → 2026-10-01
2026-10-02Verbindungswege (Quellstruktur): {"packages":[{"registryType":"npm","identifier":"@infino-ai/mcp-server","version":"0.15.0","transport":"stdio","environment":[{"name":"INFINO_MCP_URI","description":"Data to serve: a local path, an s3://|az:// bucket URI, or a hosted https://<host>/<database> endpoint (Infino Cloud). Optional — defaults to a durable per-user directory (~/.infino/mcp) so data persists across restarts, falling back to an ephemeral memory:// catalog only if that directory can't be created.","format":"string","required":false,"secret":false},{"name":"INFINO_API_KEY","description":"API key for a hosted (https://) Infino Cloud endpoint. Required when INFINO_MCP_URI is an https:// URI; ignored for local and object-storage connections.","format":"string","required":false,"secret":false},{"name":"INFINO_MCP_EMBED_PROVIDER","description":"Embedding provider: 'local' (Hugging Face transformers.js, default) or 'openai' (any OpenAI-compatible /embeddings endpoint, including Azure OpenAI's /openai/v1 surface). Inferred as 'openai' when INFINO_MCP_EMBED_BASE_URL is set.","format":"string","required":false,"secret":false},{"name":"INFINO_MCP_EMBED_BASE_URL","description":"Base URL of the OpenAI-compatible embeddings API (e.g. https://api.openai.com/v1 or https://<resource>.openai.azure.com/openai/v1). Required when the provider is 'openai'; the server POSTs to <base>/embeddings.","format":"string","required":false,"secret":false},{"name":"INFINO_MCP_EMBED_API_KEY","description":"API key for the 'openai' provider. Sent as both Authorization: Bearer and api-key so one value works for OpenAI and Azure OpenAI.","format":"string","required":false,"secret":false},{"name":"INFINO_MCP_EMBED_MODEL","description":"Embedding model. local: a Hugging Face feature-extraction model (default Xenova/all-MiniLM-L6-v2, 384-dim). openai: the model/deployment name (e.g. text-embedding-3-small, 1536-dim). Must match the model that produced the table's stored vectors.","format":"string","required":false,"secret":false},{"name":"INFINO_MCP_VALIDATE","description":"Set (1/true/yes) to probe the object store at startup, so bad credentials or an unreachable bucket fail then instead of on the first search.","format":"string","required":false,"secret":false}],"additional_arguments_declared":false}],"remotes":[]} → {"packages":[{"registryType":"npm","identifier":"@infino-ai/mcp-server","version":"0.16.0","transport":"stdio","environment":[{"name":"INFINO_MCP_URI","description":"Data to serve: a local path, an s3://|az:// bucket URI, or a hosted https://<host>/<database> endpoint (Infino Cloud). Optional — defaults to a durable per-user directory (~/.infino/mcp) so data persists across restarts, falling back to an ephemeral memory:// catalog only if that directory can't be created.","format":"string","required":false,"secret":false},{"name":"INFINO_API_KEY","description":"API key for a hosted (https://) Infino Cloud endpoint. Required when INFINO_MCP_URI is an https:// URI; ignored for local and object-storage connections.","format":"string","required":false,"secret":false},{"name":"INFINO_MCP_EMBED_PROVIDER","description":"Embedding provider: 'local' (Hugging Face transformers.js, default) or 'openai' (any OpenAI-compatible /embeddings endpoint, including Azure OpenAI's /openai/v1 surface). Inferred as 'openai' when INFINO_MCP_EMBED_BASE_URL is set.","format":"string","required":false,"secret":false},{"name":"INFINO_MCP_EMBED_BASE_URL","description":"Base URL of the OpenAI-compatible embeddings API (e.g. https://api.openai.com/v1 or https://<resource>.openai.azure.com/openai/v1). Required when the provider is 'openai'; the server POSTs to <base>/embeddings.","format":"string","required":false,"secret":false},{"name":"INFINO_MCP_EMBED_API_KEY","description":"API key for the 'openai' provider. Sent as both Authorization: Bearer and api-key so one value works for OpenAI and Azure OpenAI.","format":"string","required":false,"secret":false},{"name":"INFINO_MCP_EMBED_MODEL","description":"Embedding model. local: a Hugging Face feature-extraction model (default Xenova/all-MiniLM-L6-v2, 384-dim). openai: the model/deployment name (e.g. text-embedding-3-small, 1536-dim). Must match the model that produced the table's stored vectors.","format":"string","required":false,"secret":false},{"name":"INFINO_MCP_VALIDATE","description":"Set (1/true/yes) to probe the object store at startup, so bad credentials or an unreachable bucket fail then instead of on the first search.","format":"string","required":false,"secret":false}],"additional_arguments_declared":false}],"remotes":[]}
2026-10-02Paketversionen (Rohangabe): 0.15.0 → 0.16.0
2026-09-23Deklarierte Version: 0.14.0 → 0.15.0
2026-09-23Im Register geändert am (Rohangabe): 2026-09-08 → 2026-09-22
2026-09-23Im Register eingetragen am (Rohangabe): 2026-09-08 → 2026-09-22
2026-09-23Verbindungswege (Quellstruktur): {"packages":[{"registryType":"npm","identifier":"@infino-ai/mcp-server","version":"0.14.0","transport":"stdio","environment":[{"name":"INFINO_MCP_URI","description":"Data to serve: a local path, an s3://|az:// bucket URI, or a hosted https://<host>/<database> endpoint (Infino Cloud). Optional — defaults to a durable per-user directory (~/.infino/mcp) so data persists across restarts, falling back to an ephemeral memory:// catalog only if that directory can't be created.","format":"string","required":false,"secret":false},{"name":"INFINO_API_KEY","description":"API key for a hosted (https://) Infino Cloud endpoint. Required when INFINO_MCP_URI is an https:// URI; ignored for local and object-storage connections.","format":"string","required":false,"secret":false},{"name":"INFINO_MCP_EMBED_PROVIDER","description":"Embedding provider: 'local' (Hugging Face transformers.js, default) or 'openai' (any OpenAI-compatible /embeddings endpoint, including Azure OpenAI's /openai/v1 surface). Inferred as 'openai' when INFINO_MCP_EMBED_BASE_URL is set.","format":"string","required":false,"secret":false},{"name":"INFINO_MCP_EMBED_BASE_URL","description":"Base URL of the OpenAI-compatible embeddings API (e.g. https://api.openai.com/v1 or https://<resource>.openai.azure.com/openai/v1). Required when the provider is 'openai'; the server POSTs to <base>/embeddings.","format":"string","required":false,"secret":false},{"name":"INFINO_MCP_EMBED_API_KEY","description":"API key for the 'openai' provider. Sent as both Authorization: Bearer and api-key so one value works for OpenAI and Azure OpenAI.","format":"string","required":false,"secret":false},{"name":"INFINO_MCP_EMBED_MODEL","description":"Embedding model. local: a Hugging Face feature-extraction model (default Xenova/all-MiniLM-L6-v2, 384-dim). openai: the model/deployment name (e.g. text-embedding-3-small, 1536-dim). Must match the model that produced the table's stored vectors.","format":"string","required":false,"secret":false},{"name":"INFINO_MCP_VALIDATE","description":"Set (1/true/yes) to probe the object store at startup, so bad credentials or an unreachable bucket fail then instead of on the first search.","format":"string","required":false,"secret":false}],"additional_arguments_declared":false}],"remotes":[]} → {"packages":[{"registryType":"npm","identifier":"@infino-ai/mcp-server","version":"0.15.0","transport":"stdio","environment":[{"name":"INFINO_MCP_URI","description":"Data to serve: a local path, an s3://|az:// bucket URI, or a hosted https://<host>/<database> endpoint (Infino Cloud). Optional — defaults to a durable per-user directory (~/.infino/mcp) so data persists across restarts, falling back to an ephemeral memory:// catalog only if that directory can't be created.","format":"string","required":false,"secret":false},{"name":"INFINO_API_KEY","description":"API key for a hosted (https://) Infino Cloud endpoint. Required when INFINO_MCP_URI is an https:// URI; ignored for local and object-storage connections.","format":"string","required":false,"secret":false},{"name":"INFINO_MCP_EMBED_PROVIDER","description":"Embedding provider: 'local' (Hugging Face transformers.js, default) or 'openai' (any OpenAI-compatible /embeddings endpoint, including Azure OpenAI's /openai/v1 surface). Inferred as 'openai' when INFINO_MCP_EMBED_BASE_URL is set.","format":"string","required":false,"secret":false},{"name":"INFINO_MCP_EMBED_BASE_URL","description":"Base URL of the OpenAI-compatible embeddings API (e.g. https://api.openai.com/v1 or https://<resource>.openai.azure.com/openai/v1). Required when the provider is 'openai'; the server POSTs to <base>/embeddings.","format":"string","required":false,"secret":false},{"name":"INFINO_MCP_EMBED_API_KEY","description":"API key for the 'openai' provider. Sent as both Authorization: Bearer and api-key so one value works for OpenAI and Azure OpenAI.","format":"string","required":false,"secret":false},{"name":"INFINO_MCP_EMBED_MODEL","description":"Embedding model. local: a Hugging Face feature-extraction model (default Xenova/all-MiniLM-L6-v2, 384-dim). openai: the model/deployment name (e.g. text-embedding-3-small, 1536-dim). Must match the model that produced the table's stored vectors.","format":"string","required":false,"secret":false},{"name":"INFINO_MCP_VALIDATE","description":"Set (1/true/yes) to probe the object store at startup, so bad credentials or an unreachable bucket fail then instead of on the first search.","format":"string","required":false,"secret":false}],"additional_arguments_declared":false}],"remotes":[]}
2026-09-23Paketversionen (Rohangabe): 0.14.0 → 0.15.0
2026-09-09Deklarierte Version: 0.13.0 → 0.14.0
2026-09-09Beschreibung (Rohangabe): Keyword, vector, hybrid, and SQL retrieval over data on object storage, for AI agents. → Keyword, semantic, hybrid, and SQL retrieval for AI agents, local or on Infino Cloud.
2026-09-09Im Register geändert am (Rohangabe): 2026-09-04 → 2026-09-08
2026-09-09Im Register eingetragen am (Rohangabe): 2026-09-04 → 2026-09-08
2026-09-09Verbindungswege (Quellstruktur): {"packages":[{"registryType":"npm","identifier":"@infino-ai/mcp-server","version":"0.13.0","transport":"stdio","environment":[{"name":"INFINO_MCP_URI","description":"Data to serve: a local path, an s3://|az:// bucket URI, or a hosted https://<host>/<database> endpoint (Infino Cloud). Optional — defaults to a durable per-user directory (~/.infino/mcp) so data persists across restarts, falling back to an ephemeral memory:// catalog only if that directory can't be created.","format":"string","required":false,"secret":false},{"name":"INFINO_API_KEY","description":"API key for a hosted (https://) Infino Cloud endpoint. Required when INFINO_MCP_URI is an https:// URI; ignored for local and object-storage connections.","format":"string","required":false,"secret":false},{"name":"INFINO_MCP_EMBED_PROVIDER","description":"Embedding provider: 'local' (Hugging Face transformers.js, default) or 'openai' (any OpenAI-compatible /embeddings endpoint, including Azure OpenAI's /openai/v1 surface). Inferred as 'openai' when INFINO_MCP_EMBED_BASE_URL is set.","format":"string","required":false,"secret":false},{"name":"INFINO_MCP_EMBED_BASE_URL","description":"Base URL of the OpenAI-compatible embeddings API (e.g. https://api.openai.com/v1 or https://<resource>.openai.azure.com/openai/v1). Required when the provider is 'openai'; the server POSTs to <base>/embeddings.","format":"string","required":false,"secret":false},{"name":"INFINO_MCP_EMBED_API_KEY","description":"API key for the 'openai' provider. Sent as both Authorization: Bearer and api-key so one value works for OpenAI and Azure OpenAI.","format":"string","required":false,"secret":false},{"name":"INFINO_MCP_EMBED_MODEL","description":"Embedding model. local: a Hugging Face feature-extraction model (default Xenova/all-MiniLM-L6-v2, 384-dim). openai: the model/deployment name (e.g. text-embedding-3-small, 1536-dim). Must match the model that produced the table's stored vectors.","format":"string","required":false,"secret":false},{"name":"INFINO_MCP_VALIDATE","description":"Set (1/true/yes) to probe the object store at startup, so bad credentials or an unreachable bucket fail then instead of on the first search.","format":"string","required":false,"secret":false}],"additional_arguments_declared":false}],"remotes":[]} → {"packages":[{"registryType":"npm","identifier":"@infino-ai/mcp-server","version":"0.14.0","transport":"stdio","environment":[{"name":"INFINO_MCP_URI","description":"Data to serve: a local path, an s3://|az:// bucket URI, or a hosted https://<host>/<database> endpoint (Infino Cloud). Optional — defaults to a durable per-user directory (~/.infino/mcp) so data persists across restarts, falling back to an ephemeral memory:// catalog only if that directory can't be created.","format":"string","required":false,"secret":false},{"name":"INFINO_API_KEY","description":"API key for a hosted (https://) Infino Cloud endpoint. Required when INFINO_MCP_URI is an https:// URI; ignored for local and object-storage connections.","format":"string","required":false,"secret":false},{"name":"INFINO_MCP_EMBED_PROVIDER","description":"Embedding provider: 'local' (Hugging Face transformers.js, default) or 'openai' (any OpenAI-compatible /embeddings endpoint, including Azure OpenAI's /openai/v1 surface). Inferred as 'openai' when INFINO_MCP_EMBED_BASE_URL is set.","format":"string","required":false,"secret":false},{"name":"INFINO_MCP_EMBED_BASE_URL","description":"Base URL of the OpenAI-compatible embeddings API (e.g. https://api.openai.com/v1 or https://<resource>.openai.azure.com/openai/v1). Required when the provider is 'openai'; the server POSTs to <base>/embeddings.","format":"string","required":false,"secret":false},{"name":"INFINO_MCP_EMBED_API_KEY","description":"API key for the 'openai' provider. Sent as both Authorization: Bearer and api-key so one value works for OpenAI and Azure OpenAI.","format":"string","required":false,"secret":false},{"name":"INFINO_MCP_EMBED_MODEL","description":"Embedding model. local: a Hugging Face feature-extraction model (default Xenova/all-MiniLM-L6-v2, 384-dim). openai: the model/deployment name (e.g. text-embedding-3-small, 1536-dim). Must match the model that produced the table's stored vectors.","format":"string","required":false,"secret":false},{"name":"INFINO_MCP_VALIDATE","description":"Set (1/true/yes) to probe the object store at startup, so bad credentials or an unreachable bucket fail then instead of on the first search.","format":"string","required":false,"secret":false}],"additional_arguments_declared":false}],"remotes":[]}
2026-09-09Paketversionen (Rohangabe): 0.13.0 → 0.14.0
2026-09-08Verbindungswege (Quellstruktur): – → {"packages":[{"registryType":"npm","identifier":"@infino-ai/mcp-server","version":"0.13.0","transport":"stdio","environment":[{"name":"INFINO_MCP_URI","description":"Data to serve: a local path, an s3://|az:// bucket URI, or a hosted https://<host>/<database> endpoint (Infino Cloud). Optional — defaults to a durable per-user directory (~/.infino/mcp) so data persists across restarts, falling back to an ephemeral memory:// catalog only if that directory can't be created.","format":"string","required":false,"secret":false},{"name":"INFINO_API_KEY","description":"API key for a hosted (https://) Infino Cloud endpoint. Required when INFINO_MCP_URI is an https:// URI; ignored for local and object-storage connections.","format":"string","required":false,"secret":false},{"name":"INFINO_MCP_EMBED_PROVIDER","description":"Embedding provider: 'local' (Hugging Face transformers.js, default) or 'openai' (any OpenAI-compatible /embeddings endpoint, including Azure OpenAI's /openai/v1 surface). Inferred as 'openai' when INFINO_MCP_EMBED_BASE_URL is set.","format":"string","required":false,"secret":false},{"name":"INFINO_MCP_EMBED_BASE_URL","description":"Base URL of the OpenAI-compatible embeddings API (e.g. https://api.openai.com/v1 or https://<resource>.openai.azure.com/openai/v1). Required when the provider is 'openai'; the server POSTs to <base>/embeddings.","format":"string","required":false,"secret":false},{"name":"INFINO_MCP_EMBED_API_KEY","description":"API key for the 'openai' provider. Sent as both Authorization: Bearer and api-key so one value works for OpenAI and Azure OpenAI.","format":"string","required":false,"secret":false},{"name":"INFINO_MCP_EMBED_MODEL","description":"Embedding model. local: a Hugging Face feature-extraction model (default Xenova/all-MiniLM-L6-v2, 384-dim). openai: the model/deployment name (e.g. text-embedding-3-small, 1536-dim). Must match the model that produced the table's stored vectors.","format":"string","required":false,"secret":false},{"name":"INFINO_MCP_VALIDATE","description":"Set (1/true/yes) to probe the object store at startup, so bad credentials or an unreachable bucket fail then instead of on the first search.","format":"string","required":false,"secret":false}],"additional_arguments_declared":false}],"remotes":[]}
2026-09-05Deklarierte Version: 0.12.0 → 0.13.0
2026-09-05Im Register geändert am (Rohangabe): 2026-08-25 → 2026-09-04
2026-09-05Im Register eingetragen am (Rohangabe): 2026-08-25 → 2026-09-04
2026-09-05Beschreibungen der Umgebungsvariablen (Rohangabe): INFINO_API_KEY=API key for a hosted (https://) Infino Cloud endpoint. Required when INFINO_MCP_URI is an https:// URI; ignored for local and object-storage connections. · INFINO_MCP_EMBED_API_KEY=API key for the 'openai' provider. Sent as both Authorization: Bearer and api-key so one value works for OpenAI and Azure OpenAI. · INFINO_MCP_EMBED_BASE_URL=Base URL of the OpenAI-compatible embeddings API (e.g. https://api.openai.com/v1 or https://<resource>.openai.azure.com/openai/v1). Required when the provider is 'openai'; the server POSTs to <base>/embeddings. · INFINO_MCP_EMBED_MODEL=Embedding model. local: a Hugging Face feature-extraction model (default Xenova/all-MiniLM-L6-v2, 384-dim). openai: the model/deployment name (e.g. text-embedding-3-small, 1536-dim). Must match the model that produced the table's stored vectors. · INFINO_MCP_EMBED_PROVIDER=Embedding provider: 'local' (Hugging Face transformers.js, default) or 'openai' (any OpenAI-compatible /embeddings endpoint, including Azure OpenAI's /openai/v1 surface). Inferred as 'openai' when INFINO_MCP_EMBED_BASE_URL is set. · INFINO_MCP_ENABLE_WRITES=Set (1/true/yes) to expose add/update/delete tools and allow DDL/DML through SQL. Read-only when unset. · INFINO_MCP_URI=Data to serve: a local path, an s3://|az:// bucket URI, or a hosted https://<host>/<database> endpoint (Infino Cloud). Optional — defaults to a durable per-user directory (~/.infino/mcp) so data persists across restarts, falling back to an ephemeral memory:// catalog only if that directory can't be created. · INFINO_MCP_VALIDATE=Set (1/true/yes) to probe the object store at startup, so bad credentials or an unreachable bucket fail then instead of on the first search. → INFINO_API_KEY=API key for a hosted (https://) Infino Cloud endpoint. Required when INFINO_MCP_URI is an https:// URI; ignored for local and object-storage connections. · INFINO_MCP_EMBED_API_KEY=API key for the 'openai' provider. Sent as both Authorization: Bearer and api-key so one value works for OpenAI and Azure OpenAI. · INFINO_MCP_EMBED_BASE_URL=Base URL of the OpenAI-compatible embeddings API (e.g. https://api.openai.com/v1 or https://<resource>.openai.azure.com/openai/v1). Required when the provider is 'openai'; the server POSTs to <base>/embeddings. · INFINO_MCP_EMBED_MODEL=Embedding model. local: a Hugging Face feature-extraction model (default Xenova/all-MiniLM-L6-v2, 384-dim). openai: the model/deployment name (e.g. text-embedding-3-small, 1536-dim). Must match the model that produced the table's stored vectors. · INFINO_MCP_EMBED_PROVIDER=Embedding provider: 'local' (Hugging Face transformers.js, default) or 'openai' (any OpenAI-compatible /embeddings endpoint, including Azure OpenAI's /openai/v1 surface). Inferred as 'openai' when INFINO_MCP_EMBED_BASE_URL is set. · INFINO_MCP_URI=Data to serve: a local path, an s3://|az:// bucket URI, or a hosted https://<host>/<database> endpoint (Infino Cloud). Optional — defaults to a durable per-user directory (~/.infino/mcp) so data persists across restarts, falling back to an ephemeral memory:// catalog only if that directory can't be created. · INFINO_MCP_VALIDATE=Set (1/true/yes) to probe the object store at startup, so bad credentials or an unreachable bucket fail then instead of on the first search.
2026-09-05Paketversionen (Rohangabe): 0.12.0 → 0.13.0
2026-09-05Umgebungsvariablen (Rohangabe): INFINO_MCP_URI, INFINO_API_KEY, INFINO_MCP_ENABLE_WRITES, INFINO_MCP_EMBED_PROVIDER, INFINO_MCP_EMBED_BASE_URL, INFINO_MCP_EMBED_API_KEY, INFINO_MCP_EMBED_MODEL, INFINO_MCP_VALIDATE → INFINO_MCP_URI, INFINO_API_KEY, INFINO_MCP_EMBED_PROVIDER, INFINO_MCP_EMBED_BASE_URL, INFINO_MCP_EMBED_API_KEY, INFINO_MCP_EMBED_MODEL, INFINO_MCP_VALIDATE
2026-08-26Einsatzgebiet, aus der Anbieterbeschreibung abgeleitet: – → Datenbank
2026-08-26Beschreibung (Rohangabe): – → Keyword, vector, hybrid, and SQL retrieval over data on object storage, for AI agents.
2026-08-26Deklarierte Version: 0.10.2 → 0.12.0
2026-08-26Im Register geändert am (Rohangabe): 2026-08-24 → 2026-08-25
2026-08-26Im Register eingetragen am (Rohangabe): 2026-08-24 → 2026-08-25
2026-08-26Paketversionen (Rohangabe): 0.10.2 → 0.12.0
2026-08-25Deklarierte Version: 0.10.0 → 0.10.2
2026-08-25Im Register geändert am (Rohangabe): 2026-08-19 → 2026-08-24
2026-08-25Im Register eingetragen am (Rohangabe): 2026-08-19 → 2026-08-24
2026-08-25Paketversionen (Rohangabe): 0.10.0 → 0.10.2
2026-08-20Deklarierte Version: 0.9.0 → 0.10.0
2026-08-20Im Register geändert am (Rohangabe): 2026-08-13 → 2026-08-19
2026-08-20Im Register eingetragen am (Rohangabe): 2026-08-13 → 2026-08-19
2026-08-20Paketversionen (Rohangabe): 0.9.0 → 0.10.0
2026-08-16Im Register geändert am (Rohangabe): – → 2026-08-13
2026-08-16Im Register eingetragen am (Rohangabe): – → 2026-08-13
2026-08-16Statusmeldung des Registers (Rohangabe): – → –
2026-08-16Schemafassung des Rohsatzes (Rohangabe): – → https://static.modelcontextprotocol.io/schemas/2025-12-11/server.schema.json
2026-08-16Symbolformate (Rohangabe): – → –
2026-08-16Beschreibungen der Umgebungsvariablen (Rohangabe): – → INFINO_API_KEY=API key for a hosted (https://) Infino Cloud endpoint. Required when INFINO_MCP_URI is an https:// URI; ignored for local and object-storage connections. · INFINO_MCP_EMBED_API_KEY=API key for the 'openai' provider. Sent as both Authorization: Bearer and api-key so one value works for OpenAI and Azure OpenAI. · INFINO_MCP_EMBED_BASE_URL=Base URL of the OpenAI-compatible embeddings API (e.g. https://api.openai.com/v1 or https://<resource>.openai.azure.com/openai/v1). Required when the provider is 'openai'; the server POSTs to <base>/embeddings. · INFINO_MCP_EMBED_MODEL=Embedding model. local: a Hugging Face feature-extraction model (default Xenova/all-MiniLM-L6-v2, 384-dim). openai: the model/deployment name (e.g. text-embedding-3-small, 1536-dim). Must match the model that produced the table's stored vectors. · INFINO_MCP_EMBED_PROVIDER=Embedding provider: 'local' (Hugging Face transformers.js, default) or 'openai' (any OpenAI-compatible /embeddings endpoint, including Azure OpenAI's /openai/v1 surface). Inferred as 'openai' when INFINO_MCP_EMBED_BASE_URL is set. · INFINO_MCP_ENABLE_WRITES=Set (1/true/yes) to expose add/update/delete tools and allow DDL/DML through SQL. Read-only when unset. · INFINO_MCP_URI=Data to serve: a local path, an s3://|az:// bucket URI, or a hosted https://<host>/<database> endpoint (Infino Cloud). Optional — defaults to a durable per-user directory (~/.infino/mcp) so data persists across restarts, falling back to an ephemeral memory:// catalog only if that directory can't be created. · INFINO_MCP_VALIDATE=Set (1/true/yes) to probe the object store at startup, so bad credentials or an unreachable bucket fail then instead of on the first search.
2026-08-16Formate der Umgebungsvariablen (Rohangabe): – → string
2026-08-16Bereitstellungsform (Rohangabe): – → paket
2026-08-16Remote-Hosts (Rohangabe): – → –
2026-08-16Remote-Adressen (Rohangabe): – → –
2026-08-16Laufzeithinweise (Rohangabe): – → –
2026-08-16Paketversionen (Rohangabe): – → 0.9.0
2026-08-16Paketbezeichner (Rohangabe): – → @infino-ai/mcp-server

Diese Ansicht blättert vorwärts und nennt keine Gesamtzahl: Einzelne Zeilen entfallen erst bei der Ausgabe, wenn ihr Beleg fehlt. Eine gezählte Zahl wäre größer als die gezeigte.

tracevero · https://tracevero.de/mcp/io-github-infino-ai-mcp-server/aenderungen