Skip to main content
Glama

memory_search

Read-onlyIdempotent

Search a bounded page of the org's memory before building, when unsure about a past call, or when the user references something not in context. The same filters with page.next_cursor as cursor continue the page; pages are not drained automatically. Expand omitted fields with tasks_get view=field.

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": {
      "project_id": "<project_id>"
    },
    "name": "memory_search"
  }
}

Input Schema

NameRequiredDescriptionDefault
qNoText to search over title + description. Omitted: the whole memory is listed.
typeNoNarrow to one knowledge type (e.g. "directive"). Omitted: search covers the durable knowledge types (philosophy, directive, decision, spec, learning, principle, sop, supplier, vision, content, roadmap, epic).
limitNoMax rows to return (default 10). Compact pages return at most 20 rows; the response budget may return fewer. page.limit is the effective cap.
cursorNoOpaque page.next_cursor from the previous memory search. Its filters are repeated. Cannot be combined with offset.
offsetNoFresh row offset; cursor is the continuation path. Cannot be combined with cursor.
project_idYesThe org-memory project (UUID, subdomain, slug, or 'default'). Resolved server-side.
include_archivedNoWork records only: include archived work.
include_supersededNoInclude replaced direction and work; default false.

Schema Changelog

Changes observed during successful MCP inspections.

  1. First observed

TDQS

A4.2/5.0
Behavior4/5

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

Annotations only declare readOnlyHint and idempotentHint, so the description properly carries the interesting behavior: pages are bounded, they are not auto-drained, and continuation requires re-supplying page.next_cursor with the same filters. It also warns that the response budget may return fewer rows than the requested limit, which is non-obvious and not derivable from annotations. No contradiction with the read-only profile.

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?

Three tight sentences: triggers first, pagination mechanics second, escape hatch (tasks_get) last. Every sentence carries distinct information and the most decision-relevant content is front-loaded.

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?

With no output schema, the description covers the essentials an agent needs: how paging works, that results may be fewer than limit, and where to go for omitted fields. It does not describe the shape of a returned result row itself, leaving a modest gap for a tool with 8 parameters.

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 100%, so the schema already documents q, type, limit, cursor, offset, project_id, include_archived, and include_superseded. The description's cursor note largely restates the schema's own constraint (same filters repeated, not combinable with offset), so the baseline 3 applies with only marginal added meaning.

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?

States a specific verb ("Search") and resource ("the org's memory") and adds scope precision with "a bounded page," which immediately tells an agent this is a paged read rather than an exhaustive dump. The contrast with memory_save/tasks_get is implicit but clear from the read framing and the tasks_get pointer for field expansion.

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?

Gives three concrete invocation triggers: before building, when unsure about a past call, and when the user references something not in context — genuinely useful routing context. It also names tasks_get view=field as the path for expanding omitted fields, though it stops short of stating when NOT to use this tool (e.g. versus other search_* siblings).

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

Try in Browser

Glama MCP Gateway

Add one secure layer between your agents and this server.

Resources