Add Components

Extend your agent project with additional features.

Commands

• sekuire add memory - Add a memory backend

---

add memory

Add a memory storage backend to your agent for conversation history, state persistence, or caching.

sekuire add memory

Interactive Selection

Running the command presents a menu of memory backend options:

Terminal

$ sekuire add memory

Select memory backend for your agent:
❯ In-Memory (ephemeral)
  SQLite (local file)
  Redis (self-hosted)
  PostgreSQL
  Upstash Redis (serverless)
  Cloudflare KV
  Cloudflare D1
  AWS DynamoDB
  Turso (edge SQLite)
  Convex

Available Backends

Backend	Type	Best For
In-Memory	Ephemeral	Development, testing
SQLite	Local file	Single-instance deployments
Redis	Self-hosted	Multi-instance with existing Redis
PostgreSQL	Self-hosted	Multi-instance with existing Postgres
Upstash Redis	Serverless	Serverless deployments
Cloudflare KV	Edge	Cloudflare Workers
Cloudflare D1	Edge SQLite	Cloudflare Workers
AWS DynamoDB	Serverless	AWS deployments
Turso	Edge SQLite	Edge deployments
Convex	Serverless	Real-time applications

---

Configuration

The command updates your sekuire.yml with the memory configuration:

sekuire.yml

agent:
  memory:
    type: "sqlite"
    config:
      filename: "./agent_memory.db"

Configuration by Backend

In-Memory (Buffer)

agent:
  memory:
    type: "buffer"
    max_messages: 10

No dependencies required. Data is lost on restart.

SQLite

agent:
  memory:
    type: "sqlite"
    config:
      filename: "./agent_memory.db"

Redis

agent:
  memory:
    type: "redis"
    config:
      url: "redis://localhost:6379"

PostgreSQL

agent:
  memory:
    type: "postgres"
    config:
      url: "postgresql://localhost:5432/sekuire"

Upstash Redis

agent:
  memory:
    type: "upstash"
    config:
      url: "https://xxx.upstash.io"
      token: "${UPSTASH_REDIS_REST_TOKEN}"

Cloudflare KV

agent:
  memory:
    type: "cloudflare-kv"
    config:
      namespace_id: "your-namespace-id"
      account_id: "${CF_ACCOUNT_ID}"
      api_token: "${CF_API_TOKEN}"

Cloudflare D1

agent:
  memory:
    type: "cloudflare-d1"
    config:
      database_id: "your-database-id"
      account_id: "${CF_ACCOUNT_ID}"
      api_token: "${CF_API_TOKEN}"

AWS DynamoDB

agent:
  memory:
    type: "dynamodb"
    config:
      table_name: "sekuire-memory"
      region: "us-east-1"

Turso

agent:
  memory:
    type: "turso"
    config:
      url: "libsql://xxx.turso.io"
      auth_token: "${TURSO_AUTH_TOKEN}"

Convex

agent:
  memory:
    type: "convex"
    config:
      url: "https://xxx.convex.cloud"

---

Dependencies

Install the required package for your chosen backend:

Node.js

Backend	Package
SQLite	better-sqlite3
Redis	redis
PostgreSQL	pg
Upstash	@upstash/redis
DynamoDB	@aws-sdk/client-dynamodb, @aws-sdk/lib-dynamodb
Turso	@libsql/client
Convex	convex

Terminal

# Example: SQLite
npm install better-sqlite3

# Example: Upstash
npm install @upstash/redis

Python

Backend	Package
SQLite	(included in stdlib)
Redis	redis
PostgreSQL	psycopg2-binary
Upstash	upstash-redis
DynamoDB	boto3
Turso	libsql-experimental

Terminal

# Example: Redis
pip install redis

# Example: PostgreSQL
pip install psycopg2-binary

---

Environment Variables

Some backends require environment variables:

Backend	Variables
Upstash	UPSTASH_REDIS_REST_TOKEN
Cloudflare	CF_ACCOUNT_ID, CF_API_TOKEN
DynamoDB	AWS_ACCESS_KEY_ID, AWS_SECRET_ACCESS_KEY, AWS_REGION
Turso	TURSO_AUTH_TOKEN

---

Example Workflow

Terminal

# Add memory to existing project
$ sekuire add memory

Select memory backend for your agent:
  In-Memory (ephemeral)
❯ SQLite (local file)
  ...

SQLite database file: ./agent_memory.db

Memory backend configured successfully!
   Type: sqlite
   Config updated in sekuire.yml

Install the required dependency:
   npm install better-sqlite3
   # or
   pip install sqlite3  # (included in Python stdlib)

---

Next Steps

• init - Initialize a new project with memory built-in
• run - Run your agent locally