Favicon of engram-memory

engram-memory Skill

AI Agent SkillCode Search & MemoryPythonOpen source

engram-memory is a Python AI agent skill for Claude Desktop, Cursor, Windsurf with npx, uvx/pip, docker. The highest-scoring AI memory system ever benchmarked that isn't reliant on LLM reranking. And it's free & burns less tokens.

How to use engram-memory Skill

Install or run

npx -y install-mcp@latest http://localhost:8585/mcp \
    --client <your-client> --name engrammemory --oauth=no -y

Client configuration

{
  "name": "engrammemory",
  "display_name": "Engram Memory",
  "description": "Persistent semantic memory for AI agents. Store, search, and recall memories across sessions using a three-tier recall engine (hot cache, hash index, vector search).",
  "version": "0.2.0",
  "author": "EngramMemory",
  "author_url": "https://github.com/EngramMemory",
  "homepage": "https://engrammemory.ai",
  "repository": "https://github.com/EngramMemory/engram-memory",
  "license": "MIT",
  "icon": "https://engrammemory.ai/logo.svg",
  "privacy_policies": [
    "https://engrammemory.ai/privacy"
  ],
  "contact": {
    "email": "eddy@engrammemory.ai",
    "support_url": "https://github.com/EngramMemory/engram-memory/issues"
  },
  "tools": [
    {
      "name": "memory_store",
      "title": "Store Memory",
      "description": "Store a memory with semantic embedding",
      "readOnlyHint": false,
      "destructiveHint": false
    },
    {
      "name": "memory_search",
      "title": "Search Memories",
      "description": "Search memories using three-tier recall",
      "readOnlyHint": true,
      "destructiveHint": false
    },
    {
      "name": "memory_recall",
      "title": "Recall Context",
      "description": "Recall relevant memories for context injection",
      "readOnlyHint": true,
      "destructiveHint": false
    },
    {
      "name": "memory_forget",
      "title": "Forget Memory",
      "description": "Delete a memory from all tiers",
      "readOnlyHint": false,
      "destructiveHint": true
    },
    {
      "name": "memory_consolidate",
      "title": "Consolidate Memories",
      "description": "Find and merge near-duplicate memories",
      "readOnlyHint": false,
      "destructiveHint": true
    },
    {
      "name": "memory_feedback",
      "title": "Give Fe

Compatible clients

Claude DesktopCursorWindsurfVS CodeChatGPT

What is engram-memory Skill?

engram-memory is a Python AI agent skill for Claude Desktop, Cursor, Windsurf with npx, uvx/pip, docker. The highest-scoring AI memory system ever benchmarked that isn't reliant on LLM reranking. And it's free & burns less tokens. This profile combines repository metadata with install, compatibility, and usage signals so developers can quickly decide whether it fits their agent workflow before opening the source repository.

Trust signal
85/100
Maintenance signal
95/100
Adoption signal
38/100

Automated repository signals based on public metadata such as recency, license, installation evidence, and adoption. These are not a security audit or endorsement.

Key capabilities

  • Reusable instructions support
  • Includes SKILL.md support
  • Shell commands support
  • Presentation work support
  • Document generation support
  • OpenAI provider support
  • Anthropic provider support

Technical details

Install or run engram-memory using npx. Check EngramMemory/engram-memory for the latest setup command.

npxuvx/pipdockergit cloneremote URLClaude DesktopCursorWindsurfVS CodeChatGPTstdioHTTPSSE
  • Built for Python
  • Built for Node.js
  • Built for Rust
  • Install or run with npx
  • Install or run with uvx/pip
  • Install or run with docker
  • Install or run with git clone
  • Install or run with remote URL

Requirements and access

PythonNode.jsRustNOASSERTION licenseAPI key requiredAPI key requiredOAuthDatabase accessNetwork access

Security and permissions

Review permissions before connecting any MCP server to an agent. Pay special attention to whether it can read local files, write data, call external services, or perform destructive actions.

Destructive actionsWrite actionsDatabase accessNetwork accessCredentials requiredCan write or update dataMay support destructive actionsAPI key required authenticationHosted or remote

When to use engram-memory Skill

  • Use it for reusing agent instructions, scripts, and references.
  • Use it for agent skills workflows.
  • Use it for coding workflows.
  • Use it for data workflows.
  • Use it for claude desktop users.

Built with

PythonNode.jsRustnpxuvx/pipdockergit cloneremote URLClaude DesktopCursorWindsurfVS CodeChatGPTstdioHTTPSSE

Explore related resources

Frequently asked questions

What is engram-memory?

engram-memory is a Python AI agent skill for Claude Desktop, Cursor, Windsurf with npx, uvx/pip, docker. The highest-scoring AI memory system ever benchmarked that isn't reliant on LLM reranking. And it's free & burns less tokens.

Who is engram-memory best for?

engram-memory is best for reusing agent instructions, scripts, and references, agent skills workflows, coding workflows, data workflows, claude desktop users.

How do I install engram-memory?

Install or run engram-memory using npx. Check engram-memory for the latest setup command.

Is engram-memory actively maintained?

engram-memory may need a closer maintenance check before production use.

Share:

Stars
33
Forks
4
Last commit
2 months ago
Repository age
4 months
License
NOASSERTION

Auto-fetched from GitHub.

Similar to engram-memory