Compute embedding vectors for RAG and search using a specified model, with automatic model loading when needed.
Compute vectors with an embedding model for RAG/search. May auto-load the model.
| Name | Required | Description | Default |
|---|---|---|---|
| input | Yes | ||
| model | Yes |
Changes observed during successful MCP inspections.
v2.1.0Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations declare readOnlyHint=false, destructiveHint=false, openWorldHint=false, covering the safety profile. The description adds one genuinely useful trait not in the annotations — 'May auto-load the model' — which explains the non-read-only hint and warns of a side effect. It still omits return format, batch/size limits, and any cost or latency notes.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
Two short sentences, front-loaded with the core purpose and no wasted words. Efficient, though the brevity is partly the source of the missing detail rather than pure tightness.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Two required parameters with zero documentation, no output schema, and a thin two-sentence description leave the agent under-informed about how to call the tool or what it returns. The 'May auto-load the model' note helps but does not close the gap.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema coverage is 0% and there is no output schema, so the description carries the full burden — yet it says nothing about the 'model' or 'input' parameters. It does not clarify that model takes an identifier or that input accepts a single string or an array of strings for batch embedding.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
States a specific verb (compute) and resource (vectors) plus a purpose (RAG/search), which is enough for an agent to grasp the tool. It does not explicitly distinguish itself from related siblings like lm_rag_index or lm_rag_search, but the embedding-generating role is clear.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The phrase 'for RAG/search' implies a use context but gives no explicit when-to-use, when-not-to-use, or alternative routing. An agent cannot tell from this description whether to call lm_embeddings directly versus letting lm_rag_index or lm_rag_search produce embeddings internally.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.