marsnme

search_memories

Find past decisions and related context by asking questions in natural language. Semantic search matches meanings, not just keywords.

Instructions

Semantic search across short-term memories in demo.memories using Jina embeddings. Finds memories by meaning, not just keywords — ask a natural language question and get the most relevant matches. Use this to recall past decisions, find related context, or check if something was already discussed recently.

Tool call template

Replace each placeholder with a value matching the input schema. Only required parameters are included; add optional parameters as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {
      "query": "<query>"
    },
    "name": "search_memories"
  }
}

Input Schema

Name	Required	Description	Default
`limit`	No	Maximum number of results to return
`query`	Yes	Natural language search query. Describe what you are looking for — semantic matching finds relevant results even without exact keywords.
`scope`	No	Search scope: "this_body" searches only the current profile, "all_bodies" searches across all personas.	this_body
`source`	No
`agent_body`	No	Filter results to a specific persona/body (e.g. "coco", "toto")
`environment`	No	Filter results to a specific environment (e.g. "production", "staging")
`min_similarity`	No	Minimum cosine similarity threshold for results (range -1 to 1). Higher values return fewer but more relevant matches.
`unexpired_only`	No	When true (default), exclude expired memories from results

Schema Changelog

Changes observed during successful MCP inspections.

Changed1 schema field changedv0.1.3Jul 13, 2026

changedInput schema / properties / source / enum

Previous value: -[
-  "perplexity",
-  "cursor",
-  "warp",
-  "openclaw",
-  "hermes"
-]New value: +[
+  "perplexity",
+  "cursor",
+  "warp",
+  "openclaw",
+  "hermes",
+  "draft",
+  "grok"
+]

Changed7 schema fields changedv0.1.1Jun 10, 2026

changedInput schema / properties / agent_body / description

Previous value: -"Optional body scope key; defaults to source/body/env inference"New value: +"Filter results to a specific persona/body (e.g. \"coco\", \"toto\")"

changedInput schema / properties / environment / description

Previous value: -"Optional environment filter"New value: +"Filter results to a specific environment (e.g. \"production\", \"staging\")"

addedInput schema / properties / limit / description

Added value: +"Maximum number of results to return"

addedInput schema / properties / min_similarity / description

Added value: +"Minimum cosine similarity threshold for results (range -1 to 1). Higher values return fewer but more relevant matches."

changedInput schema / properties / query / description

Previous value: -"Semantic search query text"New value: +"Natural language search query. Describe what you are looking for — semantic matching finds relevant results even without exact keywords."

addedInput schema / properties / scope / description

Added value: +"Search scope: \"this_body\" searches only the current profile, \"all_bodies\" searches across all personas."

addedInput schema / properties / unexpired_only / description

Added value: +"When true (default), exclude expired memories from results"

First observedv0.1.0Jun 10, 2026

TDQS

A3.8/5.0

Behavior2/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

No annotations are provided, so the description must carry full behavioral burden. It states it's a semantic search but does not disclose any behavioral traits such as read-only nature, auth requirements, rate limits, or whether it modifies state. The description lacks details beyond the basic function, leaving the agent uninformed about important behavioral aspects.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness5/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The description is concise with two sentences, front-loading the key function. Every sentence adds value, and there is no unnecessary information. It is well-structured for quick understanding.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness4/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

Given 8 parameters and no output schema, the description covers the core purpose and usage. It explains semantic matching and typical use cases. However, it lacks details on result format, sorting, or performance characteristics, which would be helpful for full context. Still, it is reasonably complete for a search tool.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters3/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema description coverage is 88%, so the baseline is 3. The description adds context for the 'query' parameter (emphasizing semantic matching) but does not significantly enhance meaning for other parameters. The 'source' parameter lacks any description in the schema or description text, slightly reducing clarity.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description clearly states it's a semantic search tool for short-term memories using Jina embeddings, distinguishing it from keyword search and sibling tools like list_memories. It specifies the resource (demo.memories) and the action (semantic search by meaning), making the purpose unambiguous.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines4/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

The description provides clear context for when to use the tool: to recall past decisions, find related context, or check if something was already discussed. While it doesn't explicitly mention when not to use or alternatives, the usage hints are sufficient for an AI agent to decide.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Deploy Server