Skip to main content
Glama
jagoff

MEMO MCP Server

by jagoff

memo_save

Persist a fact, decision, or preference as a durable, searchable memory. Optionally decompose content into atomic facts with tags for finer retrieval.

Instructions

Persist content to memo.

Use memo_save for a durable curated fact, decision, or preference; use memo_offload for bulk dumps of working context.

When extract is true (defaults to the MEMO_SAVE_EXTRACT flag, off), the helper LLM decomposes content into atomic facts and saves each as its own memory (mem0 ADD-model) instead of one opaque blob; tags propagate to every fact. Returns an extraction summary (status, saved ids, saved_titles, counts) rather than a single record. If nothing extractable is found, the blob is saved verbatim. Normal saves add action (created, corroborated, or revised) and index_pending so callers can distinguish evidence from a new record.

scope controls the auto project:<repo> tag for THIS call only: "global" skips it (the memory lands untagged → the global recall tier, +0.10 boost everywhere); "project" or None keep the default auto-detection. An explicit project: tag in tags always wins either way.

defer_embed mirrors CLI memo save --defer-embed: it persists markdown + BM25 immediately and marks the semantic vector pending for memo_reindex. Extraction mode ignores it, matching the CLI.

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
tagsNoTags to attach; lower-cased and de-duplicated.
typeNoMemory type. One of: decision, fact, bug, feedback, preference, note, manual, synthesis, procedure, failure_pattern, reference, temp.note
extraNoArbitrary JSON metadata bag stored with the record. Authority-controlled keys (write_policy, visibility, trust_tier, ...) are stripped.
scopeNo'global' skips the auto project:<repo> tag (global recall tier); 'project' or None keep auto-detection. Other values are rejected.
titleNoOptional title; when omitted it is derived from the first line of content.
contentYesMarkdown body to persist. Must be non-empty.
extractNoDecompose content into atomic facts saved individually; None defers to the MEMO_SAVE_EXTRACT flag (off by default).
auto_deriveNoWhen true, a helper LLM fills missing metadata (title/type/tags); adds ~0.5-2s per save.
defer_embedNoPersist markdown and the text index immediately, but leave the semantic vector pending for a later memo_reindex call.

Output Schema

TableJSON Schema
NameRequiredDescriptionDefault

No arguments

Behavior4/5

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

Annotations only carry readOnlyHint=false and destructiveHint=false (no strong assertions), so the description must disclose behavior, which it does thoroughly. It explains the extraction mode's return shape (extraction summary vs single record), the tag propagation to every fact, the fallback 'blob saved verbatim' behavior, the auto project:<repo> tagging logic including the global recall tier boost, and the authority-controlled key stripping in `extra`. This is rich behavioral context well beyond minimal requirements.

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?

The description is organized into clear topically-paragraphed blocks (purpose, extraction mode, scope, defer_embed), each earning its place. It is long but every paragraph addresses a distinct behavioral aspect necessary for correct usage. Shorter than the length would suggest given 9 parameters and several conditional behaviors it must explain.

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?

Given 9 parameters, an output schema exists (which presumably documents the extraction summary shape), and no strong annotations to lean on, the description fully compensates. It covers the two major sub-modes (extract vs normal), the scope/recall-tier mechanics, the defer_embed flow, cross-references memo_reindex, and notes CLI parity. For a tool this complex, this is complete.

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

Parameters4/5

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

Schema coverage is 100%, so baseline is 3. The description adds meaningful value by explaining the interaction semantics between parameters: extract's decomposition behavior, scope's 'global'/None distinction and the project tag precedence ('An explicit project: tag in tags always wins'), and defer_embed's mirroring of the CLI flag with the extraction-mode exception. Parameters like title/type/auto_derive are adequately covered by schema alone.

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 the verb ('Persist `content` to memo') and resource, and explicitly distinguishes memo_save from memo_offload ('durable curated fact, decision, or preference' vs 'bulk dumps of working context'). The scope of what gets saved is precise and it differentiates from the closest sibling.

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?

The description provides explicit usage guidance: 'Use memo_save for a durable curated fact, decision, or preference; use memo_offload for bulk dumps of working context.' It also explains mode-specific behaviors (extract vs normal) and how scope/auto_derive/defer_embed each alter behavior, giving the agent clear conditional decision-making guidance.

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

Install Server

Other Tools

Latest Blog Posts

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/jagoff/memo'

If you have feedback or need assistance with the MCP directory API, please join our Discord server