Skip to main content
Glama

Server Details

Long-term memory for AI assistants. Isolated per-user storage, recall across conversations.

Ownership verified
Status
Healthy
Uptime
100.0% over the last 24 hours
OAuth
Works in Glama
Last Tested
Transport
Streamable HTTP · MCP 2025-06-18
URL
ChatGPT
–
Claude
LeapMemory

TDQS

A4.3/5.0

Scored across 6 tools

Disambiguation5/5

Each tool has a clearly distinct purpose, and the descriptions explicitly disambiguate the two closest pairs: add_rule vs remember (behavioral instruction vs fact) and briefing vs recall (standing context vs specific detail). An agent should reliably select the right tool.

Naming Consistency3/5

The rule tools follow a verb_noun snake_case pattern (add_rule, list_rules, update_rule), but briefing, recall, and remember are bare verbs/nouns, with 'briefing' being a noun among verbs. Still readable, but conventions are mixed.

Tool Count5/5

Six tools cleanly cover two subsystems (memory and rules) with no filler. Each tool earns its place: read/write for memory and create/list/update for rules.

Completeness3/5

Notable gaps: rules have add/list/update but no delete_rule, and memory has remember/recall but no update or forget/delete operation. Agents can partially work around these (e.g. update_rule to neutralize a rule), but lifecycle coverage is incomplete.

Available Tools

6 tools
add_ruleAdd a ruleAInspect

Save a standing instruction about how to work with this user, so it applies in every future conversation without being repeated. The rule text is sent to LeapMemory's servers and stored in the user's account, and the user can view, edit, and delete rules in the LeapMemory app. Use this instead of remember when the user states a preference about your behaviour, a convention, or a constraint: how to write, what never to do, how to format an answer. Facts about the user and their world still go to remember. Write the rule in the user's own words, as an instruction, not as a fact about them.

Inputs: text (required), scope.

briefingMemory briefingA
Read-only
Inspect

Returns the user's standing background summary: who they are, their preferences, the people around them, and what they have been working on most recently. Takes no arguments. Call it once when a conversation opens, before answering, to load that context; it is a session-start read, not a per-message one. Also useful when the user asks where things stand, what is next, or to continue earlier work. recall is for specific details this does not cover.

Inputs: no named parameters.

list_rulesList rulesA
Read-only
Inspect

List the user's saved rules. The always-applicable ones are already delivered at the start of the conversation, so call this only when the user asks what rules they have, or asks about rules scoped to a specific project.

Inputs: scope.

recallMemory recallA
Read-only
Inspect

Search the user's long-term memory for a specific fact, decision, project detail, or past statement. Useful before answering anything that depends on something particular the user told you earlier. For where things stand or what is in progress, call briefing instead. Returns facts and source snippets, or nothing if the memory has no answer.

Inputs: query (required).

rememberMemory saveAInspect

Save a piece of information to the user's long-term memory. The content is sent to LeapMemory's servers and stored in the user's account, so it persists across conversations and can be recalled later. The user can view and delete stored memories in the LeapMemory app. Use for durable facts the user shares about themselves, their world, or their preferences. Write the fact as a plain third-person statement about the user (e.g. 'The user prefers X'), never from your own point of view: no 'I will', 'awaiting', or references to yourself.

Inputs: content (required).

update_ruleUpdate a ruleA
Idempotent
Inspect

Change the wording or scope of a rule the user already saved. Use this when they correct or narrow an existing rule, rather than calling add_rule again, which would leave both versions in force. Call list_rules first to get the rule_id; never guess one. Pass only the fields that change.

Inputs: rule_id (required), text, scope.

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

    Related MCP Connectors

    Related MCP Servers

    • A
      license
      Not graded
      quality
      B
      maintenance
      Persistent memory for AI assistants — store insights, decisions, and knowledge that survives across conversations.
      20
      Apache 2.0
    • F
      license
      Not graded
      quality
      B
      maintenance
      Provides persistent memory for LLM applications, enabling AI assistants to remember user preferences, facts, and conversation history across sessions.
      2
      -
    • F
      license
      A
      quality
      Not graded
      maintenance
      Provides long-term memory storage for AI assistants with semantic search, enabling persistent storage of preferences, decisions, and context with relationship tracking between memories.
      19
      -
    Try in Browser

    Glama MCP Gateway

    Add one secure layer between your agents and this server.

    Resources