Skip to main content
Glama
CROW: LLM Wiki

CROW: LLM Wiki

Official

recall

Read-only

Answer questions from long-term memory with cited source notes, so you can verify details about work, documents, people, or past decisions before responding.

Instructions

Answer a question from the user's long-term memory, citing the notes the answer comes from.

Call it before you answer about the user's work, documents, people, decisions or anything from before, in any language. Not to save something: use remember; for matching notes with no answer written, search. A model writes the answer. Returns {answer, notes, titles}: notes are the cited paths, to read in full with read_note. With your models, it returns a job with requests instead: answer them with reply.

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": {
      "question": "<question>"
    },
    "name": "recall"
  }
}

Input Schema

NameRequiredDescriptionDefault
whereNoOptional: only notes with all these values, e.g. {"team": "ops"}
questionYesWhat to look up, as a question in the user's words, in any language

Output Schema

NameRequiredDescriptionDefault

No arguments

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed2 schema fields changedv1.0.5
    • changedInput schema / properties / question / description
      Previous value: -"What to look up, as a question"New value: +"What to look up, as a question in the user's words, in any language"
    • changedInput schema / properties / where / description
      Previous value: -"Only notes whose fields have all these values, e.g. {\"team\": \"ops\"}"New value: +"Optional: only notes with all these values, e.g. {\"team\": \"ops\"}"
  2. First observedv1.0.4

TDQS

A4.4/5.0
Behavior4/5

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

Beyond readOnlyHint/openWorldHint, it discloses that a model composes the answer, what fields come back ({answer, notes, titles}), that notes are cited paths to expand with `read_note`, and that a `job` with requests is returned when the caller runs its own models. That async job path is genuinely useful non-obvious behavior. It stops short of stating latency, cost, or what happens when nothing is recalled.

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

Conciseness4/5

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

Front-loads the purpose, then usage, then return shape, with no filler sentences. The final sentence is dense with the job/reply branch, but every clause carries real information.

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

Completeness5/5

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

Complete for a two-parameter read tool: purpose, trigger condition, alternatives, return fields, the follow-up call for reading cited notes, and the alternate job-based execution path are all covered, even though an output schema exists.

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 coverage is 100%, with both `question` and `where` already documented in the schema, so the baseline applies. The description adds only the 'in any language' framing already present in the schema and says nothing about the `where` filter or how question phrasing affects recall quality.

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?

Opens with a specific verb and resource: answers a question from the user's long-term memory and cites the source notes. It is immediately distinguishable from siblings because it explicitly names `remember` for saving and `search` for note matching without a written answer.

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

Usage Guidelines5/5

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

Gives an explicit trigger ('call it before you answer about the user's work, documents, people, decisions or anything from before'), an explicit exclusion ('not to save something'), and names the correct alternative for each conflicting case.

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