Qredence Documentation

fleet-rlm ships as a single FastAPI process that exposes the /api/* SSE surface described in the HTTP API reference. There is no bundled web UI; the maintained client is pi-tui.

Architecture in production

A production deploy needs:

fleet-rlm process — uv run fleet-rlm serve-api (equivalent to uv run fleet web).
Daytona — Sandbox provider (FLEET_DAYTONA_API_KEY).
Postgres at the canonical Alembic head — FLEET_DATABASE_URL.
LLM provider — OpenCode Go (interactive profiles) or Databricks AI Gateway (managed profile).
MLflow (optional) — tracing backend. The daytona-managed profile requires managed Unity Catalog inputs.

Fleet has no caller authentication. Backend launchers default to binding 127.0.0.1 and reject non-loopback hosts unless --allow-non-loopback-bind is passed. In production, terminate TLS and enforce access control at a reverse proxy in front of Fleet on a private interface.

Environment configuration

Select the runtime policy in config/fleet.toml:

[config]
default_profile = "daytona-managed"

Then set only the variables named by that profile:

.env.production

FLEET_DAYTONA_API_KEY=...
FLEET_DATABASE_URL=postgresql+asyncpg://USER:PASS@HOST/DB?sslmode=require

# daytona-managed / benchmark profiles
DATABRICKS_TOKEN=...
FLEET_DATABRICKS_AI_GATEWAY_BASE_URL=https://<gateway>/serving-endpoints
FLEET_MLFLOW_EXPERIMENT_NAME=...
FLEET_MLFLOW_TRACE_CATALOG=...
FLEET_MLFLOW_TRACE_SCHEMA=...
FLEET_MLFLOW_TRACE_TABLE_PREFIX=...
FLEET_MLFLOW_TRACING_SQL_WAREHOUSE_ID=...

See the configuration reference for the complete matrix.

Initialize the database

Fleet never applies migrations at startup. Bring the database to the canonical Alembic head before starting the backend:

uv run python scripts/db_init.py
uv run alembic check

Run the backend

uv run fleet-rlm serve-api --port 8000

To bind a non-loopback interface behind a proxy, pass the deliberate opt-in:

uv run fleet-rlm serve-api --host 0.0.0.0 --port 8000 --allow-non-loopback-bind

The /api/settings endpoint still rejects non-loopback clients when the main API is exposed. Route settings edits through pi-tui /settings on a loopback client.

Health probes

Two unauthenticated endpoints are designed for orchestrators and load balancers:

Endpoint	Purpose
`GET /health`	Liveness.
`GET /ready`	Readiness including planner, database, and sandbox provider status. Returns `503` with the same `ReadyResponse` body when a critical dependency is missing or degraded.

Reverse proxy notes

Terminate TLS at the proxy and forward to Fleet on a private interface.
Disable response buffering. The Turn stream is Server-Sent Events; buffered proxies will delay data-status heartbeats and Runtime Events.
The Turn stream emits a transient data-status chunk every runtime.heartbeat_seconds while preparation resolves. Set the proxy read timeout above that heartbeat.

Verify the deploy

Run the Daytona doctor before load-testing a real Turn:

uv run fleet doctor daytona

Then hit the readiness endpoint:

curl https://your-deploy.example.com/ready

Fleet-RLM

Fleet Pi

GEPA Omni

Qredence Skills

Fleet Reasoner

Qredence Plugins

Deploy the fleet-rlm backend

Architecture in production

Environment configuration

Initialize the database

Run the backend

Health probes

Reverse proxy notes

Verify the deploy

See also

​Architecture in production

​Environment configuration

​Initialize the database

​Run the backend

​Health probes

​Reverse proxy notes

​Verify the deploy

​See also

Architecture in production

Environment configuration

Initialize the database

Run the backend

Health probes

Reverse proxy notes

Verify the deploy

See also