Shodh-Memory AI Agent Memory

MCPOfficialOpen SourceApache-2.070.8

by varun29ankuS • Uncategorized

An AI agent memory system that persistently remembers important information, forgets irrelevant data, and improves over time without external dependencies.

Example Use Cases

1
Persistent memory across sessions without relying on cloud or external APIs
2
Fast, offline semantic search and memory recall
3
A self-improving memory system that learns from usage and forgets irrelevant data

Description

Shodh-Memory provides persistent, intelligent memory for AI agents by using algorithmic intelligence such as local embeddings, decay, and learned associations instead of relying on LLM API calls. It runs fully offline as a single lightweight binary, enabling fast memory storage and recall without cloud or external databases. The system improves with use by strengthening frequently accessed memories and fading irrelevant ones, supporting better context retention across sessions.

Capabilities(48 total)

Tools (38)

remember

Store a memory for future recall. Use this to remember important information, decisions, user preferences, project context, or anything you want to recall later.

content:string*type:stringtags:arraycreated_at:stringemotional_valence:numberemotional_arousal:numberemotion:stringsource_type:stringcredibility:numberepisode_id:stringsequence_number:numberpreceding_memory_id:stringparent_id:stringimportance:numberrobot_id:stringmission_id:stringgeo_location:arraylocal_position:arrayheading:numberaction_type:stringreward:numbersensor_data:objectoutcome_type:stringterrain_type:string

recall

Search memories AND todos using semantic similarity. Returns both relevant memories and matching todos. Use this to find past experiences, decisions, context, or pending work. Modes: 'semantic' (vector similarity), 'associative' (graph traversal), 'temporal' (time-based retrieval), 'hybrid' (combined), 'spatial' (geo-location based), 'mission' (mission context), 'action_outcome' (reward-based learning).

query:string*limit:numbermode:stringsession_id:stringrobot_id:stringmission_id:stringgeo_lat:numbergeo_lon:numbergeo_radius_meters:numberaction_type:stringreward_min:numberreward_max:numberoutcome_type:stringfailures_only:booleanterrain_type:stringtags:array

recall_by_tags

Find memories by tags. Returns memories matching ANY of the provided tags. Useful for finding memories by category (e.g., 'tool:Edit', 'file:src/main.rs', 'source:hook', 'error', 'session-summary').

tags:array*limit:number

context_summary

Get a condensed summary of recent learnings, decisions, and context. Use this at the start of a session to quickly understand what you've learned before.

include_decisions:booleaninclude_learnings:booleaninclude_context:booleanmax_items:number

list_memories

List all stored memories

limit:number

Prompts (6)

quick_recall

Search your memories for relevant context

query

session_summary

Get a summary of recent learnings, decisions, and context

what_i_know

Surface everything related to a topic

topic

pending_work

Show todos and incomplete tasks

recent_memories

Show recently created memories

count

Resources (4)

Available Commands

List all shodh-memory commands and their usage

URI: shodh://commandsMIME: text/markdown

Session Summary

Recent learnings, decisions, and context

URI: shodh://summaryMIME: text/plain

Pending Work

Your todo list and incomplete tasks

URI: shodh://todosMIME: text/plain

Memory Stats

Memory system statistics and health

URI: shodh://statsMIME: application/json

Quick Actions

View on GitHub

Security

Scanned 1 month(s) ago

Risk Level

MINIMAL

Read-only data retrieval, no side effects

Trust Score

C63/100

8/17 checks passed

Scores are informational only and provided “as is” without warranty. AgentHotspot assumes no liability for actions taken based on these ratings.

Quick Stats

Service TypeMCP

Pricing ModelFree

Capabilities38 Tools / 6 Prompts / 4 Resources

Ownervarun29ankuS

CategoryUncategorized

DependenciesStandalone

Set Your Username