Schema | Seahorse RAG MCP

Seahorse RAG MCP

Describes the environment variables required to run the server.

Name	Required	Description	Default
`GLINER_MODEL`	No	Default GLiNER fallback model id (Hugging Face).	fastino/gliner2-multi-v1
`RERANK_RRF_K`	No	RRF constant k.	60
`RERANK_TOP_N`	No	Number of top PPR candidates passed to fusion.	20
`RERANK_W_PPR`	No	PPR fusion weight (auto-normalized with rerank weight).	0.7
`HIPPO_NODE_MAX`	No	Override profile's max nodes limit.	None
`RERANK_ENABLED`	No	Enable local cross-encoder reranker.	False
`RERANK_W_RERANK`	No	Reranker fusion weight (auto-normalized with PPR weight).	0.3
`GLINER_ONNX_PATH`	No	Local ONNX GLiNER bundle directory.	models/gliner_onnx
`RERANK_MODEL_PATH`	No	Path to reranker ONNX model (.onnx).	None
`HIPPO_MEMORY_ALPHA`	No	Override memory allocation fraction (e.g. 0.1).	None
`GLINER_ONNX_REPO_ID`	No	ONNX GLiNER repo id for auto/setup download.	lmo3/gliner2-multi-v1-onnx
`RERANK_FUSION_METHOD`	No	weighted_sum or rrf.	weighted_sum
`RERANK_MODEL_MAX_LEN`	No	Cross-encoder max input length.	512
`RERANK_QUERY_MAX_LEN`	No	Query token cap before packing.	64
`RERANK_TOKENIZER_PATH`	No	Path to reranker tokenizer.json.	None
`GLINER_ONNX_AUTO_SETUP`	No	Auto-download ONNX GLiNER when local bundle is missing.	True
`RERANK_PASSAGE_MAX_LEN`	No	Passage token cap before packing.	448
`HIPPO_PERFORMANCE_PROFILE`	No	auto, low, mid, or high.	auto

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.

Server capabilities have not been inspected yet.

Functions exposed to the LLM to take actions

Name	Description
No tools

Interactive templates invoked by user choice

Name	Description
No prompts

Contextual data attached and managed by the client

Name	Description
No resources

This server cannot be deployed

ActivityInactive

ResponsivenessNo issues