Engram

Name: Engram
Author: tstockham96

by tstockham96communityv0.5.1

Engram is an intelligence layer for AI agents that stores memories, learns patterns, detects contradictions, and surfaces context. It provides a memory solution that goes beyond simple storage by investing intelligence at read time.

From the registry: Intelligent agent memory with automatic extraction, consolidation, and bi-temporal recall.

GitHub Repository

Install & configure

Prerequisites

required
Install engram-sdk globally
Install the engram-sdk npm package globally.
```
$ npm install -g engram-sdk
```
required
Run engram init
Run one-time initialization to set up MCP integration with Claude Code / Cursor and configure the database.
```
$ engram init
```
required
Gemini API key (or alternative LLM)
LLM API key for embeddings and consolidation. Works with Gemini, OpenAI, Ollama, Groq, Cerebras.

Install in your client

Paste into the AI you're chatting with — it will figure out the rest

Please install the `engram-tstockham96` MCP server into my current AI client (that's you).

Required prerequisites (do these first if not already done):
- **Install engram-sdk globally** — Install the engram-sdk npm package globally. Run: `npm install -g engram-sdk`
- **Run engram init** — Run one-time initialization to set up MCP integration with Claude Code / Cursor and configure the database. Run: `engram init`
- **Gemini API key (or alternative LLM)** — LLM API key for embeddings and consolidation. Works with Gemini, OpenAI, Ollama, Groq, Cerebras.

Canonical MCP server config (stdio transport):
- command: `engram`
- args: ["mcp"]
- required environment variables:
  - `GEMINI_API_KEY`: Gemini API key for embeddings and consolidation (or use ENGRAM_LLM_BASE_URL for alternative providers). (example: `<your-gemini-api-key>`)
- optional environment variables:
  - `ENGRAM_LLM_BASE_URL`: Custom API base URL (Groq, Cerebras, Ollama, etc.). (example: `<your-llm-base-url>`)
  - `ENGRAM_LLM_MODEL`: LLM model name. (example: `<your-llm-model-name>`)
  - `ENGRAM_DB_PATH`: SQLite database path. (example: `~/.engram/default.db`)
  - `ENGRAM_AUTH_TOKEN`: Bearer token for API auth. (example: `<your-auth-token>`)

Note: README says run 'engram init' to set up MCP; the specific stdio command may be auto-registered by init. Requires an LLM API key (Gemini default). 10 memory tools.

Add this MCP server to my current client's config in the correct format for you. If you need secrets or credentials I haven't provided, ASK me — do not invent values or leave raw placeholders. After adding it, tell me how to verify the server is connected.

Note: README says run 'engram init' to set up MCP; the specific stdio command may be auto-registered by init. Requires an LLM API key (Gemini default). 10 memory tools.