Redis Tasks¶

Redis-backed task board for multi-replica production deployments. Required for agent teams running across multiple gateway instances.

Configuration¶

Server-Side (Gateway Config)¶

providers:
  tasks:
    backend: "agentic_primitives_gateway.primitives.tasks.redis.RedisTasksProvider"
    config:
      redis_url: "redis://localhost:6379/0"

Parameter	Default	Description
`redis_url`	`redis://localhost:6379/0`	Redis connection URL

Environment Variables¶

Variable	Default	Description
`REDIS_URL`	`redis://localhost:6379/0`	Redis connection URL

Running Redis Locally¶

docker run -d -p 6379:6379 redis:7-alpine

Using with Agent Teams¶

The tasks primitive is used internally by the team runner. Configure it alongside your team:

providers:
  tasks:
    default: "redis"
    backends:
      redis:
        backend: "agentic_primitives_gateway.primitives.tasks.redis.RedisTasksProvider"
        config:
          redis_url: "${REDIS_URL:=redis://localhost:6379/0}"

teams:
  store:
    backend: redis
    config:
      redis_url: "${REDIS_URL:=redis://localhost:6379/0}"
  specs:
    research-team:
      planner: "planner"
      synthesizer: "synthesizer"
      workers: ["researcher", "coder"]

How It Works¶

Storage: each team run's tasks are stored as a Redis hash (tasks:{team_run_id}). Fields are task IDs, values are JSON-serialized task objects.
Atomic claiming: a Lua script checks task status + dependency completion and sets claimed in a single round-trip. No distributed locks needed.
Status updates: task status changes and notes are also atomic Lua-scripted operations.
Multi-replica safe: multiple gateway replicas can claim and execute tasks concurrently without conflicts.

When to Use¶

Production deployments with multiple gateway replicas
Agent teams that need shared task state
Any deployment where task persistence across restarts is required

For single-replica development, use the In-Memory provider instead.

Prerequisites¶

pip install agentic-primitives-gateway[redis]
Running Redis instance (v5.0+)