Environment Variables

Complete reference for all Aurora environment variables. Configuration is done via the .env file in the project root.

Quick Setup

# Create .env from template
cp .env.example .env

# Or use the init script (auto-generates secrets)
make init

Core Configuration

Application Environment

Variable	Default	Description
AURORA_ENV	dev	Environment mode: dev, staging, production

Database

PostgreSQL connection settings.

Variable	Default	Description
POSTGRES_USER	aurora	Database username
POSTGRES_PASSWORD	required	Database password (auto-generated by make init)
POSTGRES_DB	aurora_db	Database name
POSTGRES_HOST	postgres	Database host (container name in Docker)
POSTGRES_PORT	5432	Database port

POSTGRES_USER=aurora
POSTGRES_PASSWORD=your-secure-64-char-password
POSTGRES_DB=aurora_db
POSTGRES_HOST=postgres
POSTGRES_PORT=5432

Concurrency & Connection Pool

Tuning parameters for gunicorn, Celery, and the database connection pool. Defaults work for small deployments; increase for higher concurrency.

Variable	Default	Description
GUNICORN_WORKERS	2	Number of gunicorn worker processes
GUNICORN_THREADS	32	Threads per gunicorn worker (total parallel requests = workers x threads)
CELERY_CONCURRENCY	4	Number of concurrent Celery task workers
DB_POOL_MIN	2	Minimum database connections kept open per process
DB_POOL_MAX	20	Maximum database connections per process; excess threads wait briefly for a connection

GUNICORN_WORKERS=2
GUNICORN_THREADS=32
CELERY_CONCURRENCY=4
DB_POOL_MIN=2
DB_POOL_MAX=20

For higher concurrency, scale these together. Example for a 4-vCPU pod:

GUNICORN_WORKERS=4
GUNICORN_THREADS=8
DB_POOL_MAX=40
CELERY_CONCURRENCY=8

On Kubernetes, prefer horizontal pod autoscaling over large per-pod thread counts. See the Kubernetes deployment guide for HPA configuration.

Redis

Redis connection for Celery task queue and caching.

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

REDIS_URL=redis://redis:6379/0

Object Storage

S3-compatible object storage configuration. SeaweedFS is included by default.

Variable	Default	Description
STORAGE_BUCKET	aurora-storage	Bucket name
STORAGE_ENDPOINT_URL	http://seaweedfs-filer:8333	S3 endpoint URL
STORAGE_ACCESS_KEY	admin	Access key
STORAGE_SECRET_KEY	admin	Secret key
STORAGE_REGION	us-east-1	AWS region (for S3 compatibility)
STORAGE_USE_SSL	false	Use HTTPS for storage
STORAGE_VERIFY_SSL	false	Verify SSL certificates
STORAGE_CACHE_ENABLED	true	Enable storage response caching
STORAGE_CACHE_TTL	60	Cache TTL in seconds

Default (SeaweedFS)

STORAGE_BUCKET=aurora-storage
STORAGE_ENDPOINT_URL=http://seaweedfs-filer:8333
STORAGE_ACCESS_KEY=admin
STORAGE_SECRET_KEY=admin
STORAGE_REGION=us-east-1
STORAGE_USE_SSL=false

AWS S3

STORAGE_BUCKET=your-bucket-name
STORAGE_ENDPOINT_URL=https://s3.amazonaws.com
STORAGE_ACCESS_KEY=AKIAXXXXXXXXXXXXXXXX
STORAGE_SECRET_KEY=your-secret-key
STORAGE_REGION=us-east-1
STORAGE_USE_SSL=true
STORAGE_VERIFY_SSL=true

Cloudflare R2

STORAGE_BUCKET=your-bucket-name
STORAGE_ENDPOINT_URL=https://accountid.r2.cloudflarestorage.com
STORAGE_ACCESS_KEY=your-access-key
STORAGE_SECRET_KEY=your-secret-key
STORAGE_REGION=auto
STORAGE_USE_SSL=true

URLs & Networking

Service URLs for internal and external communication.

Variable	Default	Description
FRONTEND_URL	http://localhost:3000	Public frontend URL
BACKEND_URL	http://aurora-server:5080	Internal backend URL (container-to-container)
NEXT_PUBLIC_BACKEND_URL	http://localhost:5080	Public backend URL (browser access)
NEXT_PUBLIC_WEBSOCKET_URL	ws://localhost:5006	Public WebSocket URL
CHATBOT_INTERNAL_URL	http://chatbot:5007	Internal chatbot URL

FRONTEND_URL=http://localhost:3000
BACKEND_URL=http://aurora-server:5080
NEXT_PUBLIC_BACKEND_URL=http://localhost:5080
NEXT_PUBLIC_WEBSOCKET_URL=ws://localhost:5006
CHATBOT_INTERNAL_URL=http://chatbot:5007

Remote / VM Deployment

When deploying on a remote server or VM, set the NEXT_PUBLIC_* and FRONTEND_URL variables to the machine's IP or hostname:

FRONTEND_URL=http://YOUR_IP:3000
NEXT_PUBLIC_BACKEND_URL=http://YOUR_IP:5080
NEXT_PUBLIC_WEBSOCKET_URL=ws://YOUR_IP:5006

These values are injected at container startup (not baked at build time), so you do not need to rebuild images after changing them. Just recreate the frontend container:

docker compose -f docker-compose.prod-local.yml up -d frontend

Authentication & Security

Variable	Default	Description
FLASK_SECRET_KEY	required	Flask session secret (auto-generated by make init)
FLASK_PORT	5080	Flask server port
AUTH_SECRET	required	Auth.js secret for session encryption
INTERNAL_API_SECRET	(empty)	Shared secret between Next.js and Flask. When set, Flask rejects requests without a valid X-Internal-Secret header. Auto-generated by make init.
AUTH_URL	${FRONTEND_URL}	Auth.js canonical URL (set automatically in docker-compose from FRONTEND_URL)

FLASK_SECRET_KEY=your-secure-64-char-secret
FLASK_PORT=5080
AUTH_SECRET=your-secure-64-char-secret
INTERNAL_API_SECRET=your-secure-64-char-secret
# AUTH_URL is set from FRONTEND_URL in docker-compose — no need to set manually

Vault (Secrets Management)

HashiCorp Vault configuration for storing user credentials.

Variable	Default	Description
VAULT_ADDR	http://vault:8200	Vault server address
VAULT_TOKEN	required	Vault authentication token
VAULT_KV_MOUNT	aurora	KV secrets engine mount path
VAULT_KV_BASE_PATH	users	Base path for user secrets

VAULT_ADDR=http://vault:8200
VAULT_TOKEN=hvs.your-vault-token
VAULT_KV_MOUNT=aurora
VAULT_KV_BASE_PATH=users

Getting Vault Token

On first startup, get the root token from:

docker logs vault-init 2>&1 | grep "Root Token:"

LLM & AI Services

At least one LLM provider API key is required. See LLM Providers for detailed setup instructions.

Variable	Default	Description
OPENROUTER_API_KEY	-	OpenRouter API key (recommended)
OPENAI_API_KEY	-	OpenAI API key
ANTHROPIC_API_KEY	-	Anthropic API key
GOOGLE_AI_API_KEY	-	Google AI API key
LLM_PROVIDER_MODE	openrouter	Provider routing mode (see below)
AGENT_RECURSION_LIMIT	240	Max agent reasoning steps
INVESTIGATION_TIMEOUT_SECONDS	3600	Wall-clock limit for one investigation, in seconds. Default is one hour. Clamped to 120–10800. Requires a process restart. On Kubernetes the chart raises the worker termination grace and the nginx proxy timeouts to at least this value.

LLM_PROVIDER_MODE

Controls how Aurora routes LLM requests:

Mode	Description	Required key
openrouter	All requests go through OpenRouter. One key gives access to models from Anthropic, OpenAI, Google, and others.	OPENROUTER_API_KEY
direct	Requests go directly to each provider's API based on the model prefix (e.g. anthropic/... → Anthropic API). No OpenRouter account needed, but you need a separate API key for each provider you use.	Provider-specific key(s)
auto	Same behaviour as direct.	Provider-specific key(s)
provider name	Set to a provider such as bedrock, vertex, anthropic, openai, google, or ollama to route every model selection through that provider (clean model picks are translated to its native id). Models the provider can't serve fall back to their own native provider.	That provider's config

openrouter is recommended for most deployments — a single key, broadest model selection, and no need to manage multiple provider accounts.

# Use one of these
OPENROUTER_API_KEY=sk-or-v1-your-key
OPENAI_API_KEY=sk-your-key
ANTHROPIC_API_KEY=sk-ant-your-key
GOOGLE_AI_API_KEY=your-key

LLM_PROVIDER_MODE=openrouter   # or: direct
AGENT_RECURSION_LIMIT=240
INVESTIGATION_TIMEOUT_SECONDS=3600

Vertex AI (Google Cloud)

Variable	Default	Description
VERTEX_AI_PROJECT	-	Google Cloud project ID
VERTEX_AI_LOCATION	global	Vertex AI location
VERTEX_AI_SERVICE_ACCOUNT_JSON	-	Service account key JSON string

VERTEX_AI_PROJECT=my-gcp-project
VERTEX_AI_LOCATION=global
VERTEX_AI_SERVICE_ACCOUNT_JSON={"type":"service_account",...}
LLM_PROVIDER_MODE=vertex   # route every model pick through Vertex

Ollama (Local Models)

Variable	Default	Description
OLLAMA_BASE_URL	http://host.docker.internal:11434	Ollama server URL

OLLAMA_BASE_URL=http://host.docker.internal:11434

AWS Bedrock

One bedrock provider with two modes, auto-selected: set BEDROCK_BASE_URL for gateway mode (OpenAI-compatible endpoint in front of Bedrock), or leave it unset for native mode (AWS SDK). BEDROCK_* variables take precedence over the standard AWS_* ones.

For native mode, set LLM_PROVIDER_MODE=bedrock to route clean model picks (e.g. anthropic/claude-sonnet-5) through Bedrock automatically — translated to the matching inference-profile id, region-aware. For gateway mode (or to pin specific ids), use LLM_PROVIDER_MODE=direct and point MAIN_MODEL (etc.) at an explicit bedrock/<id> model.

Variable	Default	Description
BEDROCK_BASE_URL	-	Gateway mode: OpenAI-compatible base URL (e.g. .../v1). When set, Aurora uses gateway mode.
BEDROCK_API_KEY	not-needed	Gateway mode only. Optional — many VPC gateways need no key.
BEDROCK_REGION	-	Native mode AWS region. Falls back to AWS_REGION / AWS_DEFAULT_REGION. Set to global to use Bedrock's global. inference profiles, which bill at the direct list price instead of the ~10% cross-region premium.
BEDROCK_ACCESS_KEY_ID	(AWS_ACCESS_KEY_ID)	Native mode access key. Omit to use an IAM role / default credential chain.
BEDROCK_SECRET_ACCESS_KEY	(AWS_SECRET_ACCESS_KEY)	Native mode secret key.
BEDROCK_SESSION_TOKEN	(AWS_SESSION_TOKEN)	Native mode session token, used alongside BEDROCK_ACCESS_KEY_ID / BEDROCK_SECRET_ACCESS_KEY for temporary / STS credentials. Omit for long-lived keys, a profile, or an IAM role.
BEDROCK_PROFILE	-	Native mode named AWS profile (alternative to explicit keys).

# Gateway mode (OpenAI-compatible endpoint, e.g. a Bedrock Access Gateway in your VPC)
BEDROCK_BASE_URL=https://bedrock-gateway.internal.example.com/v1
BEDROCK_API_KEY=                       # optional
LLM_PROVIDER_MODE=direct
MAIN_MODEL=bedrock/anthropic.claude-3-5-sonnet-20240620-v1:0

# Native mode (AWS SDK) — leave BEDROCK_BASE_URL unset
BEDROCK_REGION=us-east-1
BEDROCK_ACCESS_KEY_ID=AKIA...          # or use BEDROCK_PROFILE / an IAM role
BEDROCK_SECRET_ACCESS_KEY=...
BEDROCK_SESSION_TOKEN=                  # omit unless using temporary / STS credentials
LLM_PROVIDER_MODE=bedrock               # routes clean model picks through Bedrock
MAIN_MODEL=anthropic/claude-sonnet-5    # auto-translated to us.anthropic.claude-sonnet-5

Web Search

Variable	Default	Description
SEARXNG_URL	http://searxng:8080	SearXNG internal URL
SEARXNG_SECRET	-	SearXNG secret key

AI Features

Variable	Default	Description
RCA_MODEL	-	Model for background RCA (format: provider/model). Overrides RCA_OPTIMIZE_COSTS when set. Default RCA path when ORCHESTRATOR_ENABLED=false.
MEMORY_SELECTOR_MODEL	(provider fast model)	Model that picks which memories to inject. It has about 6 seconds. The loader used to return only the newest 200 memories. Unset uses a fast model the active provider can serve: Gemini 3.5 Flash-Lite (Google/OpenRouter/Vertex), Haiku 5.5 (Anthropic and LLM_PROVIDER_MODE=bedrock), or GPT-5.6 Luna (OpenAI). Ollama uses MAIN_MODEL when that id is ollama/.... A Bedrock gateway (direct mode, MAIN_MODEL=bedrock/<id>) stays on that id; pin a fast one with bedrock/<id>.
JIRA_RCA_JUDGE_MODEL	RCA_MODEL	Yes/no model for Jira automatic RCA. Unset uses the resolved RCA_MODEL.
RCA_OPTIMIZE_COSTS	true	Only used when RCA_MODEL is not set. true = anthropic/claude-haiku-5.5, false = anthropic/claude-opus-5.5
ORCHESTRATOR_ENABLED	false	Multi-agent RCA orchestrator (opt-in). When true, a lead orchestrator triages each background RCA and may fan out parallel read-only sub-agents; RCA_MODEL is ignored. When false (default), RCA uses the single-agent path with RCA_MODEL.
RCA_ORCHESTRATOR_MODEL	-	Only when ORCHESTRATOR_ENABLED=true. Brain model for triage + synthesis. Format: provider/model.
RCA_SUBAGENT_MODEL	-	Only when ORCHESTRATOR_ENABLED=true. Sub-agent investigator model. Per-role overrides in orchestrator/roles/*.md frontmatter take precedence. Format: provider/model.
GEMINI_DISABLE_THINKING	-	Disable Gemini thinking mode

AI Safety Guardrails

Three-layer safety: input rail + signature matcher + LLM safety judge. See Command Safety for details.

Variable	Default	Description
GUARDRAILS_ENABLED	true	Master switch. When enabled (default), all three layers run and every LLM check fails closed on error. Set to false to disable all guardrails.
GUARDRAILS_LLM_MODEL	(MAIN_MODEL)	Model used by the safety judge and input rail. Same format and routing as MAIN_MODEL.
GUARDRAILS_DISABLE_THINKING	true	Shorten the safety judge. Gemini 3.6 Flash uses minimal; Gemini 3.8 Flash, Pro, and Argon use low (minimal is rejected). Gemini 2.5 Flash uses a budget of 0; 2.5 Pro uses 128. GPT-5 and the o-series use reasoning effort low; gpt-5.5-pro uses medium. Haiku 4.5 and 5.5 get thinking disabled (Opus rejects that field). Ollama and OpenRouter only get it for weights that advertise a think switch (qwen3, deepseek-r1, qwq, gpt-oss, magistral); a plain Llama is left alone because Ollama returns 400 otherwise. Set to false to leave the model on its default. Requires a restart.
GUARDRAILS_LLM_TIMEOUT_SECONDS	10	Seconds the safety judge may run before the command is blocked. Clamped to 5–60.
GUARDRAILS_SIGMA_ENABLED	true	Gates the vendored SigmaHQ rule corpus inside the signature matcher. Requires GUARDRAILS_ENABLED=true. Set to false to run only hand-written rules.

Cloud Providers

GCP (Google Cloud Platform)

Variable	Description
CLIENT_ID	GCP OAuth Client ID
CLIENT_SECRET	GCP OAuth Client Secret

CLIENT_ID=your-client-id.apps.googleusercontent.com
CLIENT_SECRET=your-client-secret

AWS (Amazon Web Services)

Variable	Default	Description
AWS_ACCESS_KEY_ID	-	Aurora's AWS access key
AWS_SECRET_ACCESS_KEY	-	Aurora's AWS secret key
AWS_DEFAULT_REGION	us-east-1	Default AWS region

AWS_ACCESS_KEY_ID=AKIAXXXXXXXXXXXXXXXX
AWS_SECRET_ACCESS_KEY=your-secret-key
AWS_DEFAULT_REGION=us-east-1

note

These are Aurora's own AWS credentials for STS AssumeRole calls, not end-user credentials.

Cloud Provider Cache

Variable	Default	Description
AURORA_SETUP_CACHE_ENABLED	true	Cache cloud provider setup data
AURORA_SETUP_CACHE_TTL	3600	Cache TTL in seconds
AURORA_VERIFY_CLI_IDENTITY	false	Verify CLI identity
AURORA_CACHE_TOKEN_IN_REDIS	false	Cache tokens in Redis

Third-Party Integrations

GitHub

GitHub App is the default auth path. OAuth is a flag-gated fallback for on-prem deployments that cannot expose a public webhook URL.

Variable	Default	Description
GITHUB_AUTH_MODE	app	One of app | oauth | hybrid. Controls which CTA the connector dialog renders and which auth paths the backend accepts.
GITHUB_APP_ID		Numeric App ID from the App settings page.
GITHUB_APP_CLIENT_ID		App's Client ID (starts with Iv23l...).
NEXT_PUBLIC_GITHUB_APP_SLUG		The App's URL slug (e.g. aurora-acme). Used by the frontend to build install management URLs.
GITHUB_APP_WEBHOOK_URL		Public URL Aurora exposes for webhooks. Must match what's configured on the App. Example: https://aurora.example.com/github/webhook.
GITHUB_APP_SETUP_URL		Post-install redirect URL. Example: https://aurora.example.com/github/app/install/callback.
GITHUB_APP_WEBHOOK_SECRET		Fallback only — the secrets backend path aurora/system/github-app/webhook-secret (Vault or AWS Secrets Manager) takes precedence.
GH_OAUTH_CLIENT_ID		OAuth App Client ID. Required only when GITHUB_AUTH_MODE is oauth or hybrid.
GH_OAUTH_CLIENT_SECRET		OAuth App Client Secret. Required only when GITHUB_AUTH_MODE is oauth or hybrid.
HOSTED_SIGNUP_ENABLED	false	Hosted deployments only. Enables the one-click website signup flow (/github/app/signup/start): installing the GitHub App provisions an Aurora org + admin account from the GitHub identity and auto-enrolls the granted repos in Incident Prevention. Leave off for self-hosted installs (registration stays invite-only).
GITHUB_APP_CLIENT_SECRET		Client secret from the App settings page. Required only when HOSTED_SIGNUP_ENABLED=true — used to exchange the install-time OAuth code for the installer's GitHub identity. The App must have "Request user authorization (OAuth) during installation" enabled and https://<aurora-host>/github/app/signup/callback registered as a Callback URL.
NEXT_PUBLIC_ENABLE_INCIDENT_PREVENTION	true	Master switch for pre-merge PR risk review (GitHub and Bitbucket). Must be set on the server and the Celery worker — the webhook ingress and the review task each check it independently.

App-mode (recommended):

GITHUB_AUTH_MODE=app
GITHUB_APP_ID=12345
GITHUB_APP_CLIENT_ID=Iv23liExampleClientId
NEXT_PUBLIC_GITHUB_APP_SLUG=aurora-acme
GITHUB_APP_WEBHOOK_URL=https://aurora.example.com/github/webhook
GITHUB_APP_SETUP_URL=https://aurora.example.com/github/app/install/callback
GITHUB_APP_WEBHOOK_SECRET=
# (private key PEM lives in your secrets backend — Vault or AWS Secrets
#  Manager — at aurora/system/github-app/private-key, not in .env)

Hybrid (App + OAuth, e.g. for a migration window):

GITHUB_AUTH_MODE=hybrid
# all GITHUB_APP_* vars from the App-mode block above, plus:
GH_OAUTH_CLIENT_ID=your-oauth-client-id
GH_OAUTH_CLIENT_SECRET=your-oauth-client-secret

OAuth-only (on-prem fallback when public webhook ingress is unavailable):

GITHUB_AUTH_MODE=oauth
GH_OAUTH_CLIENT_ID=your-oauth-client-id
GH_OAUTH_CLIENT_SECRET=your-oauth-client-secret

See the GitHub connector setup walkthroughs for the full operator click-through.

Bitbucket

Bitbucket connects via a scoped API token entered in the UI (no env vars) or optional OAuth (BB_OAUTH_CLIENT_ID / BB_OAUTH_CLIENT_SECRET + NEXT_PUBLIC_ENABLE_BITBUCKET_OAUTH=true).

Incident Prevention on Bitbucket needs no additional environment variables: connect Bitbucket, flip the per-repo Incident Prevention toggle, and paste the returned webhook URL ({API_URL}/bitbucket/webhook/{org_id}) + org secret into each repository's Bitbucket webhook settings (events pullrequest:created and pullrequest:updated). The per-org HMAC secret is generated by Aurora and stored in the secrets backend, not in .env. The shared NEXT_PUBLIC_ENABLE_INCIDENT_PREVENTION flag (see the GitHub table above) gates this feature too.

Aurora can also create, verify and remove that webhook for you if the connected credentials carry the read:webhook:bitbucket, write:webhook:bitbucket and delete:webhook:bitbucket scopes (the single Webhooks scope on an OAuth consumer covers all three). Without them the manual paste above still works — the hook is confirmed automatically on the first pull request event, since Bitbucket sends no test event when a webhook is saved. Note that without delete:webhook:bitbucket, disabling Incident Prevention stops all PR reviews but leaves the webhook in place, so Bitbucket keeps sending events that Aurora discards; remove it by hand to stop them.

Optional: to post reviews as a dedicated bot account instead of the connected user, provision a system secret (again, not an env var):

vault kv put aurora/system/bitbucket-bot/credentials \
  value='{"email": "aurora-bot@example.com", "api_token": "..."}'

See the Bitbucket connector guide for the full setup walkthrough.

Slack

Variable	Description
SLACK_CLIENT_ID	Slack App Client ID
SLACK_CLIENT_SECRET	Slack App Client Secret
SLACK_SIGNING_SECRET	Slack App Signing Secret
SLACK_APP_TOKEN	App-level token (xapp-...) with connections:write. Set it to enable Socket Mode (outbound WebSocket) instead of HTTP webhooks — for private/self-hosted deployments with no public URL. Leave empty to use HTTP webhooks.

SLACK_CLIENT_ID=your-client-id
SLACK_CLIENT_SECRET=your-client-secret
SLACK_SIGNING_SECRET=your-signing-secret

# Private/self-hosted only (no public webhook URL) — see the Slack connector docs
SLACK_APP_TOKEN=

Google Chat

Variable	Description
GOOGLE_CHAT_CLIENT_ID	OAuth client ID (used during setup to create the incidents space in the customer's workspace)
GOOGLE_CHAT_CLIENT_SECRET	OAuth client secret
GOOGLE_CHAT_SERVICE_ACCOUNT_KEY	JSON key for the Chat app service account (messages appear as "Aurora")

GOOGLE_CHAT_CLIENT_ID=your-client-id
GOOGLE_CHAT_CLIENT_SECRET=your-client-secret
GOOGLE_CHAT_SERVICE_ACCOUNT_KEY='{"type":"service_account",...}'

PagerDuty

Variable	Default	Description
NEXT_PUBLIC_ENABLE_PAGERDUTY_OAUTH	false	Enable PagerDuty OAuth in UI
PAGERDUTY_CLIENT_ID	-	PagerDuty OAuth Client ID
PAGERDUTY_CLIENT_SECRET	-	PagerDuty OAuth Client Secret

NEXT_PUBLIC_ENABLE_PAGERDUTY_OAUTH=true
PAGERDUTY_CLIENT_ID=your-client-id
PAGERDUTY_CLIENT_SECRET=your-client-secret

OAuth now requests incidents.write (for posting RCA notes); register it on the PagerDuty app, and existing OAuth connections must disconnect and reconnect to receive it.

OVH

Variable	Default	Description
NEXT_PUBLIC_ENABLE_OVH	false	Enable OVH connector in UI
OVH_EU_CLIENT_ID	-	OVH EU region Client ID
OVH_EU_CLIENT_SECRET	-	OVH EU region Client Secret
OVH_CA_CLIENT_ID	-	OVH CA region Client ID
OVH_CA_CLIENT_SECRET	-	OVH CA region Client Secret
OVH_US_CLIENT_ID	-	OVH US region Client ID
OVH_US_CLIENT_SECRET	-	OVH US region Client Secret

NEXT_PUBLIC_ENABLE_OVH=true
OVH_EU_CLIENT_ID=your-eu-client-id
OVH_EU_CLIENT_SECRET=your-eu-client-secret

Scaleway

The Scaleway connector is always available. Users connect with an API key; no server-side credentials are required.

Confluence

Variable	Description
CONFLUENCE_CLIENT_ID	Confluence OAuth Client ID
CONFLUENCE_CLIENT_SECRET	Confluence OAuth Client Secret

CONFLUENCE_CLIENT_ID=your-client-id
CONFLUENCE_CLIENT_SECRET=your-client-secret

SharePoint

Variable	Default	Description
NEXT_PUBLIC_ENABLE_SHAREPOINT	false	Enable SharePoint connector in UI
SHAREPOINT_CLIENT_ID	-	Azure App Client ID
SHAREPOINT_CLIENT_SECRET	-	Azure App Client Secret
SHAREPOINT_TENANT_ID	common	Azure Tenant ID

NEXT_PUBLIC_ENABLE_SHAREPOINT=true
SHAREPOINT_CLIENT_ID=your-client-id
SHAREPOINT_CLIENT_SECRET=your-client-secret
SHAREPOINT_TENANT_ID=your-tenant-id

Email (SMTP)

Variable	Default	Description
SMTP_HOST	-	SMTP server hostname
SMTP_PORT	587	SMTP server port
SMTP_USER	-	SMTP username
SMTP_PASSWORD	-	SMTP password
SMTP_FROM_EMAIL	-	From email address
SMTP_FROM_NAME	Aurora	From display name

SMTP_HOST=smtp.example.com
SMTP_PORT=587
SMTP_USER=your-username
SMTP_PASSWORD=your-password
SMTP_FROM_EMAIL=aurora@example.com
SMTP_FROM_NAME=Aurora

Kubernetes & Pod Isolation

Variable	Default	Description
ENABLE_POD_ISOLATION	false	Enable Kubernetes pod isolation
TERMINAL_NAMESPACE	-	Namespace for terminal pods
TERMINAL_IMAGE	-	Container image for terminals
TERMINAL_POD_TTL	-	Pod time-to-live
TERMINAL_RUNTIME_CLASS	-	RuntimeClass for pods
CHATBOT_POD_TTL	-	Chatbot pod TTL
USE_UNTRUSTED_NODES	-	Allow untrusted nodes
NEXT_PUBLIC_KUBECTL_AGENT_CHART_URL	-	Helm chart URL for kubectl agent

MCP Server

Variable	Default	Description
MCP_PORT	8811	Port the MCP server listens on
BACKEND_URL	http://aurora-server:5080	Internal Aurora API URL (shared with other services)

The MCP server authenticates clients via Bearer tokens stored in the mcp_tokens database table. No additional environment variable is needed for token configuration -- see the MCP integration guide for details.

Custom MCP Servers

Controls Aurora connecting out to MCP servers you register, which is the opposite direction from the section above. See Custom MCP Servers.

Variable	Default	Description
MCP_ALLOW_PRIVATE_TARGETS	true (Compose, Helm, and when unset)	Allow registering MCP servers on private, loopback or link-local addresses. Default is on for self-hosted OSS. Set false on multi-tenant SaaS: with it on, anyone holding connectors:write can aim Aurora at any address it can reach, including cloud metadata endpoints.

The redirect URI Aurora presents to OAuth providers is derived from FRONTEND_URL as <FRONTEND_URL>/mcp/callback. Credentials go to the configured secrets backend, so there is no env var for them.

Rate Limiting

Variable	Default	Description
RATE_LIMITING_ENABLED	false	Enable rate limiting
RATE_LIMIT_BYPASS_TOKEN	-	Token to bypass rate limits
RATE_LIMIT_HEADERS_ENABLED	true	Include rate limit headers in responses

RATE_LIMITING_ENABLED=true
RATE_LIMIT_BYPASS_TOKEN=your-bypass-token
RATE_LIMIT_HEADERS_ENABLED=true

Development

Variable	Default	Description
NGROK_URL	-	ngrok tunnel URL for local development

Environment File Locations

File	Purpose
.env	Your local configuration (gitignored)
.env.example	Template with all variables and documentation

Docker Compose Files

File	Purpose	Gunicorn mode
docker-compose.yaml	Development stack	--reload (auto-restart on code changes)
docker-compose.prod-local.yml	Production-like local deployment	--preload (load app before forking workers)
docker-compose.airtight.yml	Air-gapped deployment from pre-built images	--preload

All three files read GUNICORN_WORKERS, GUNICORN_THREADS, CELERY_CONCURRENCY, DB_POOL_MIN, and DB_POOL_MAX from the .env file. Set them there to tune concurrency across all environments.

Keep Docker Compose Files in Sync

When adding new environment variables, update all three Docker Compose files to ensure consistency.