Environment Variables

All configuration options for OpenSRE can be set via environment variables. This page provides a complete reference.

LLM Providers

Variable	Default	Description	Module
`LLM_PROVIDER`	`anthropic`	Primary LLM provider (`anthropic`, `openai`, `openrouter`, `deepseek`, `gemini`, `nvidia`, `minimax`, `groq`, `azure-openai`, `bedrock`, `ollama`, `codex`, `claude-code`, `cursor`, `gemini-cli`, `antigravity-cli`, `opencode`, `kimi`, `copilot`, `grok-cli`, `pi`)	`config`
`LLM_MAX_TOKENS`	`4096`	Maximum tokens for LLM responses	`config`
`OPENSRE_LLM_TRANSPORT`	`sdk`	LLM transport for API providers: `sdk` (native SDKs) or `litellm`	`core/llm/transport_mode`

Anthropic

Variable	Description	Module
`ANTHROPIC_API_KEY`	API key for Anthropic	`llm_credentials`
`ANTHROPIC_REASONING_MODEL`	Model for reasoning tasks	`config`
`ANTHROPIC_TOOLCALL_MODEL`	Model for tool calling	`config`

OpenAI

Variable	Description	Module
`OPENAI_API_KEY`	API key for OpenAI	`llm_credentials`
`OPENAI_REASONING_MODEL`	Model for reasoning tasks	`config`
`OPENAI_TOOLCALL_MODEL`	Model for tool calling	`config`

OpenRouter

Variable	Description	Module
`OPENROUTER_API_KEY`	API key for OpenRouter	`llm_credentials`
`OPENROUTER_MODEL`	Shared model for all tasks	`config`
`OPENROUTER_REASONING_MODEL`	Override model for reasoning	`config`
`OPENROUTER_TOOLCALL_MODEL`	Override model for tool calling	`config`

DeepSeek

Variable	Description	Module
`DEEPSEEK_API_KEY`	API key for DeepSeek	`llm_credentials`
`DEEPSEEK_MODEL`	Shared model for all tasks	`config`
`DEEPSEEK_REASONING_MODEL`	Override model for reasoning	`config`
`DEEPSEEK_TOOLCALL_MODEL`	Override model for tool calling	`config`

Google Gemini

Variable	Description	Module
`GEMINI_API_KEY`	API key for Google Gemini	`llm_credentials`
`GEMINI_MODEL`	Shared model for all tasks	`config`
`GEMINI_REASONING_MODEL`	Override model for reasoning	`config`
`GEMINI_TOOLCALL_MODEL`	Override model for tool calling	`config`

NVIDIA

Variable	Description	Module
`NVIDIA_API_KEY`	API key for NVIDIA NIM	`llm_credentials`
`NVIDIA_MODEL`	Shared model for all tasks	`config`
`NVIDIA_REASONING_MODEL`	Override model for reasoning	`config`
`NVIDIA_TOOLCALL_MODEL`	Override model for tool calling	`config`

MiniMax

Variable	Description	Module
`MINIMAX_API_KEY`	API key for MiniMax	`llm_credentials`
`MINIMAX_MODEL`	Shared model for all tasks	`config`
`MINIMAX_REASONING_MODEL`	Override model for reasoning	`config`
`MINIMAX_TOOLCALL_MODEL`	Override model for tool calling	`config`

Groq

Variable	Description	Module
`GROQ_API_KEY`	API key for Groq	`llm_credentials`
`GROQ_MODEL`	Shared model for all tasks	`config`
`GROQ_REASONING_MODEL`	Override model for reasoning	`config`
`GROQ_TOOLCALL_MODEL`	Override model for tool calling	`config`

Azure OpenAI

Variable	Default	Description	Module
`AZURE_OPENAI_BASE_URL`		Azure OpenAI resource URL (`https://<resource>.openai.azure.com`)	`config`
`AZURE_OPENAI_API_KEY`		API key for Azure OpenAI	`llm_credentials`
`AZURE_OPENAI_API_VERSION`	`2024-10-21`	Azure OpenAI API version	`config`
`AZURE_OPENAI_MODEL`		Shared deployment name for all slots	`config`
`AZURE_OPENAI_REASONING_MODEL`	`gpt-5.4-mini`	Deployment for reasoning tasks	`config`
`AZURE_OPENAI_CLASSIFICATION_MODEL`	`gpt-5.4-mini`	Deployment for classification	`config`
`AZURE_OPENAI_TOOLCALL_MODEL`	`gpt-5.4-mini`	Deployment for tool calling	`config`

Amazon Bedrock

Variable	Default	Description	Module
`BEDROCK_REASONING_MODEL`		Model for reasoning tasks	`config`
`BEDROCK_TOOLCALL_MODEL`		Model for tool calling	`config`

Ollama (Local)

Variable	Default	Description	Module
`OLLAMA_MODEL`		Model name in Ollama	`config`
`OLLAMA_HOST`	`http://localhost:11434`	Ollama server URL	`config`

LLM Reasoning Effort

Variable	Default	Description	Module
`OPENSRE_REASONING_EFFORT`	`medium`	Reasoning effort level for extended thinking models (`low`, `medium`, `high`)	`llm_reasoning_effort`

CLI Investigation Tools

Variable	Description	Module
`CODEX_BIN`	Path to Codex CLI binary	`config`
`CODEX_MODEL`	Model override for Codex	`config`
`CLAUDE_CODE_BIN`	Path to Claude Code CLI binary	`config`
`CLAUDE_CODE_MODEL`	Model override for Claude Code	`config`
`GEMINI_CLI_BIN`	Path to Gemini CLI binary	`config`
`GEMINI_CLI_MODEL`	Model override for Gemini CLI	`config`
`ANTIGRAVITY_CLI_BIN`	Path to Antigravity CLI binary	`config`
`ANTIGRAVITY_CLI_MODEL`	Model override for Antigravity	`config`
`ANTIGRAVITY_CLI_TIMEOUT_SECONDS`	Timeout for Antigravity CLI invocations	`config`
`OPENCODE_BIN`	Path to OpenCode CLI binary	`config`
`OPENCODE_MODEL`	Model override for OpenCode	`config`
`CURSOR_BIN`	Path to Cursor CLI binary	`config`
`CURSOR_MODEL`	Model override for Cursor	`config`
`KIMI_BIN`	Path to Kimi CLI binary	`config`
`KIMI_MODEL`	Model override for Kimi	`config`
`KIMI_API_KEY`	API key for Kimi	`config`
`COPILOT_BIN`	Path to GitHub Copilot CLI binary	`config`
`COPILOT_MODEL`	Model override for Copilot	`config`
`COPILOT_HOME`	Copilot config directory override	`config`
`COPILOT_GITHUB_TOKEN`	GitHub token for Copilot auth	`config`
`GROK_CLI_BIN`	Path to Grok Build CLI binary	`config`
`GROK_CLI_MODEL`	Model override for Grok Build CLI	`config`
`XAI_API_KEY`	API-key auth for Grok Build CLI	`integrations/llm_cli`
`PI_BIN`	Path to Pi CLI binary	`config`
`PI_MODEL`	Model override for Pi CLI (`provider/model`)	`config`
`PI_AGENT_DIR`	Pi agent config directory override	`integrations/llm_cli`
`PI_CONFIG_DIR`	Pi config directory override	`integrations/llm_cli`

Output & Debugging

Variable	Default	Description	Module
`TRACER_VERBOSE`	`0`	Verbose output for tracer (`1`, `true`, `yes`)	`surfaces/interactive_shell/ui/output`
`TRACER_OUTPUT_FORMAT`		Output format for tracer results	`platform/observability/output_format`
`TRACER_API_URL`		API URL for tracer reports	`integrations/slack/delivery.py`
`NO_COLOR`		Disable colored output (standard convention)	`platform/observability/output_format`
`COLUMNS`	`80`	Terminal column width for output formatting	`surfaces/interactive_shell/ui/output`
`ENV`	`development`	Environment (`development`, `production`)	`config`

Credentials & Authentication

Sensitive variables (*_TOKEN, *_KEY, *_PASSWORD, *_SECRET, and similar) resolve through resolve_env_credential: process env first, then the OS keyring, then the local fallback store described below. Wizard setup dual-writes many of these so a cleared .env still works locally. Exceptions (never stored): webhook and other *_URL values such as SLACK_WEBHOOK_URL and ROCKETCHAT_WEBHOOK_URL stay store / plain env only. They may still embed secret tokens — do not log them unmasked. Set OPENSRE_DISABLE_KEYRING=1 to skip local credential storage entirely; you then have to export every credential in your shell.

When there is no system keychain

Headless Linux, EC2 over SSH, Docker, and CI have no keyring backend, and a macOS login Keychain can be locked. Rather than failing setup, OpenSRE writes the credential to ~/.opensre/credentials.json with owner-only (0600) permissions and tells you it did. Unlock or install a keychain and re-run the command to move the credential back into secure storage. To keep credentials off disk entirely, set OPENSRE_DISABLE_KEYRING=1 and export them in your shell instead. Contributor contract: Credential resolution in Adding tools and integrations (internal contributor markdown in the repo; not a Mintlify docs-site page).

Variable	Default	Description	Module
`OPENSRE_DISABLE_KEYRING`	`0`	Disable local credential storage; credentials must come from the environment	`config/secrets`
`OPENSRE_LLM_AUTH_METADATA_PATH`	`~/.opensre/llm-auth.json`	Non-secret LLM auth metadata path used by prompt-safe status checks	`llm_auth/records`
`OPENSRE_API_KEY`		API key for remote OpenSRE instances	`remote/server`
`DBUS_SESSION_BUS_ADDRESS`		D-Bus socket address (Linux keyring)	`llm_credentials`

Telemetry & Monitoring

Variable	Default	Description	Module
`OPENSRE_NO_TELEMETRY`	`0`	Disable all telemetry, including Sentry and PostHog (covers `$ai_generation` prompt/response events)	`platform/observability/sentry_sdk.py`
`OPENSRE_SENTRY_DISABLED`	`0`	Disable Sentry error reporting	`platform/observability/sentry_sdk.py`
`OPENSRE_SENTRY_DSN`		Override Sentry DSN	`platform/observability/sentry_sdk.py`
`OPENSRE_SENTRY_LOGGING_DISABLED`	`0`	Disable Sentry log integration	`.env.example`
`OPENSRE_DEPLOYMENT_METHOD`	`local`	Deployment method tag for Sentry (`railway`, `ec2`, `vercel`, `local`)	`.env.example`
`OPENSRE_PROMPT_LOG_DISABLED`	`0`	Disable prompt logging entirely	`.env.example`
`OPENSRE_PROMPT_LOG_LOCAL_DISABLED`	`0`	Disable local prompt log file	`.env.example`
`OPENSRE_PROMPT_LOG_REDACT`	`1`	Redact known token shapes from prompts/responses before local or PostHog logging	`.env.example`
`OPENSRE_PROMPT_LOG_PATH`	`~/.config/opensre/prompt_log.jsonl`	Path to local prompt log file	`.env.example`
`DO_NOT_TRACK`	`0`	Honor global do-not-track preference	`platform/observability/sentry_sdk.py`
`SENTRY_TRACES_SAMPLE_RATE`	`1.0`	Trace sampling rate for Sentry	`.env.example`
`SENTRY_ERROR_SAMPLE_RATE`	`1.0`	Error sampling rate for Sentry	`.env.example`

Paths & Directories

Variable	Default	Description	Module
`INVESTIGATIONS_DIR`	`~/.opensre/investigations`	Directory for saved investigations	`remote/server`
`OPENSRE_PROJECT_ENV_PATH`	`PROJECT_ROOT/.env`	Path to project-level `.env` file	`surfaces/cli/wizard/config`
`OPENSRE_WIZARD_STORE_PATH`	`~/.opensre/opensre.json`	Path to the local wizard/provider selection store	`config/constants`

Feature Flags

Variable	Default	Description	Module
`OSRE_HELM_INTEGRATION`		Enable Helm integration (1/true/yes)	`integrations/_catalog_impl`
`GITLAB_MR_WRITEBACK`		Enable GitLab merge request writeback (`true`, `1`, `yes`)	`tools/investigation/reporting/gitlab_writeback`
`OPENSRE_RELEASES_API_URL`	GitHub releases API	Override releases API endpoint	`surfaces/cli/lifecycle/update`

Integration Credentials

Observability & Monitoring

Variable	Default	Description	Module
`GRAFANA_INSTANCE_URL`		Grafana instance URL	`integrations/_catalog_impl`
`GRAFANA_READ_TOKEN`		Grafana API token	`integrations/_catalog_impl`
`GRAFANA_LOKI_DATASOURCE_UID`		UID of Loki datasource in Grafana	`.env.example`
`GRAFANA_TEMPO_DATASOURCE_UID`		UID of Tempo datasource in Grafana	`.env.example`
`GRAFANA_VERIFY_SSL`	`true`	Verify Grafana SSL certificate	`integrations/_catalog_impl`
`GRAFANA_CA_BUNDLE`		Path to a CA bundle for Grafana TLS verification (self-signed/internal CA)	`integrations/_catalog_impl`
`DD_API_KEY`		Datadog API key	`integrations/_catalog_impl`
`DD_APP_KEY`		Datadog application key	`integrations/_catalog_impl`
`DD_SITE`	`datadoghq.com`	Datadog site (us/eu)	`integrations/_catalog_impl`
`GROUNDCOVER_API_KEY`		groundcover read-only service-account token (alias: `GROUNDCOVER_MCP_TOKEN`)	`integrations/_catalog_impl`
`GROUNDCOVER_MCP_URL`	`https://mcp.groundcover.com/api/mcp`	groundcover MCP endpoint	`integrations/_catalog_impl`
`GROUNDCOVER_TENANT_UUID`		Tenant UUID (multi-workspace accounts)	`integrations/_catalog_impl`
`GROUNDCOVER_BACKEND_ID`		Backend ID (multi-backend tenants)	`integrations/_catalog_impl`
`GROUNDCOVER_TIMEZONE`	`UTC`	Timezone for returned timestamps (`X-Timezone`)	`integrations/_catalog_impl`
`HONEYCOMB_API_KEY`		Honeycomb API key	`integrations/_catalog_impl`
`HONEYCOMB_DATASET`	`__all__`	Honeycomb dataset slug	`integrations/_catalog_impl`
`HONEYCOMB_API_URL`	`https://api.honeycomb.io`	Honeycomb API endpoint	`integrations/_catalog_impl`
`CORALOGIX_API_KEY`		Coralogix API key	`integrations/_catalog_impl`
`CORALOGIX_API_URL`	`https://api.coralogix.com`	Coralogix API endpoint	`integrations/_catalog_impl`
`CORALOGIX_APPLICATION_NAME`		Application name in Coralogix	`integrations/_catalog_impl`
`CORALOGIX_SUBSYSTEM_NAME`		Subsystem name in Coralogix	`integrations/_catalog_impl`
`SIGNOZ_URL`		SigNoz instance URL	`.env.example`
`SIGNOZ_API_KEY`		SigNoz API key	`.env.example`
`ALERTMANAGER_URL`		Alertmanager URL	`integrations/_catalog_impl`
`ALERTMANAGER_BEARER_TOKEN`		Bearer token for Alertmanager	`integrations/_catalog_impl`
`ALERTMANAGER_USERNAME`		Username for Alertmanager basic auth	`integrations/_catalog_impl`
`ALERTMANAGER_PASSWORD`		Password for Alertmanager basic auth	`integrations/_catalog_impl`
`SPLUNK_URL`		Splunk instance URL	`integrations/_catalog_impl`
`SPLUNK_TOKEN`		Splunk API token	`integrations/_catalog_impl`
`SPLUNK_INDEX`	`main`	Splunk index	`integrations/_catalog_impl`
`SPLUNK_VERIFY_SSL`	`true`	Verify Splunk SSL certificate	`integrations/_catalog_impl`
`SENTRY_URL`	`https://sentry.io`	Sentry instance URL	`integrations/_catalog_impl`
`SENTRY_ORG_SLUG`		Sentry organization slug	`integrations/_catalog_impl`
`SENTRY_PROJECT_SLUG`		Sentry project slug	`integrations/_catalog_impl`
`SENTRY_AUTH_TOKEN`		Sentry auth token	`integrations/_catalog_impl`
`POSTHOG_PROJECT_ID`		PostHog project ID	`integrations/posthog`
`POSTHOG_PERSONAL_API_KEY`		PostHog personal API key	`integrations/posthog`
`POSTHOG_BASE_URL`	`https://us.i.posthog.com`	PostHog API instance base URL	`integrations/posthog`
`POSTHOG_TIMEOUT_SECONDS`	`15.0`	Timeout in seconds for PostHog API requests	`integrations/posthog`

Cloud Platforms

Variable	Default	Description	Module
`AWS_REGION`	`us-east-1`	AWS region	`integrations/_catalog_impl`
`AWS_ROLE_ARN`		AWS role ARN for cross-account access	`integrations/_catalog_impl`
`AWS_EXTERNAL_ID`		External ID for AWS cross-account role	`integrations/_catalog_impl`
`AWS_ACCESS_KEY_ID`		AWS access key	`integrations/_catalog_impl`
`AWS_SECRET_ACCESS_KEY`		AWS secret access key	`integrations/_catalog_impl`
`AWS_SESSION_TOKEN`		AWS session token	`integrations/_catalog_impl`

Messaging

Variable	Description	Module
`SLACK_WEBHOOK_URL`	Slack incoming webhook URL	`interactive_shell/ui/output`, `integrations/slack/delivery.py`
`SLACK_BOT_TOKEN`	Slack bot token	`scheduler/credentials`, `.env.example`
`SLACK_ACCESS_TOKEN`	Slack user token	`scheduler/credentials`
`SLACK_GITHUB_ISSUES_WEBHOOK_URL`	Slack webhook for GitHub issues	`integrations/github/issue_comments`
`DISCORD_BOT_TOKEN`	Discord bot token	`scheduler/credentials`, `remote/server`
`DISCORD_APPLICATION_ID`	Discord application ID	`integrations/_catalog_impl`
`DISCORD_PUBLIC_KEY`	Discord public key	`integrations/_catalog_impl`
`DISCORD_DEFAULT_CHANNEL_ID`	Default Discord channel	`integrations/_catalog_impl`
`TELEGRAM_BOT_TOKEN`	Telegram bot token	`integrations/telegram/credentials`, `scheduler/credentials`
`TELEGRAM_DEFAULT_CHAT_ID`	Default Telegram chat ID	`integrations/telegram/credentials`, `integrations/_catalog_impl`
`ROCKETCHAT_SERVER_URL`	Rocket.Chat server base URL	`integrations/_catalog_impl`
`ROCKETCHAT_AUTH_TOKEN`	Rocket.Chat personal access token	`integrations/_catalog_impl`
`ROCKETCHAT_USER_ID`	Rocket.Chat user ID for the token	`integrations/_catalog_impl`
`ROCKETCHAT_DEFAULT_CHANNEL`	Default Rocket.Chat channel	`integrations/_catalog_impl`
`ROCKETCHAT_WEBHOOK_URL`	Rocket.Chat incoming webhook URL	`integrations/_catalog_impl`
`TWILIO_ACCOUNT_SID`	Twilio account SID	`integrations/_catalog_impl`
`TWILIO_AUTH_TOKEN`	Twilio auth token	`integrations/_catalog_impl`
`TWILIO_WHATSAPP_FROM`	WhatsApp sender number (Twilio)	`integrations/_catalog_impl`
`WHATSAPP_DEFAULT_TO`	Default WhatsApp recipient	`integrations/_catalog_impl`
`TWILIO_SMS_FROM`	SMS sender number (Twilio)	`integrations/_catalog_impl`
`TWILIO_SMS_MESSAGING_SERVICE_SID`	SMS messaging service SID	`integrations/_catalog_impl`
`TWILIO_SMS_DEFAULT_TO`	Default SMS recipient	`integrations/_catalog_impl`

Code & Collaboration

Variable	Default	Description	Module
`GITHUB_MCP_MODE`	`streamable-http`	GitHub MCP transport mode	`integrations/_catalog_impl`
`GITHUB_MCP_URL`		GitHub MCP server URL	`integrations/_catalog_impl`
`GITHUB_MCP_COMMAND`		GitHub MCP command to execute	`integrations/_catalog_impl`
`GITHUB_MCP_ARGS`		GitHub MCP command arguments	`integrations/_catalog_impl`
`GITHUB_MCP_AUTH_TOKEN`		GitHub MCP auth token	`integrations/_catalog_impl`
`GITHUB_MCP_TOOLSETS`	`repos,issues,pull_requests,actions,search`	Enabled GitHub MCP toolsets	`integrations/_catalog_impl`
`OPENSRE_GITHUB_MCP_REPO_PROBE_LIMIT`	`50`	Max repos to probe (5-500)	`.env.example`
`GITLAB_BASE_URL`		GitLab instance URL	`integrations/_catalog_impl`
`GITLAB_TOKEN`		GitLab API token	`.env.example`
`JIRA_BASE_URL`		Jira instance URL	`integrations/_catalog_impl`
`JIRA_EMAIL`		Jira user email	`integrations/_catalog_impl`
`JIRA_API_TOKEN`		Jira API token	`integrations/_catalog_impl`
`JIRA_PROJECT_KEY`		Default Jira project key	`integrations/_catalog_impl`
`BITBUCKET_WORKSPACE`		Bitbucket workspace name	`integrations/_catalog_impl`
`BITBUCKET_USERNAME`		Bitbucket username	`integrations/_catalog_impl`
`BITBUCKET_APP_PASSWORD`		Bitbucket app password	`integrations/_catalog_impl`
`BITBUCKET_MAX_RESULTS`	`25`	Max results per API call	`integrations/_catalog_impl`
`INCIDENT_IO_BASE_URL`		incident.io API URL	`integrations/_catalog_impl`
`INCIDENT_IO_API_TOKEN`		incident.io API token	`.env.example`
`OPSGENIE_API_KEY`		Opsgenie API key	`integrations/_catalog_impl`
`OPSGENIE_REGION`	`us`	Opsgenie region (us/eu)	`integrations/_catalog_impl`
`VERCEL_API_TOKEN`		Vercel API token	`integrations/_catalog_impl`
`VERCEL_TEAM_ID`		Vercel team ID	`integrations/_catalog_impl`
`VERCEL_POLL_PROJECT_IDS`		Project IDs for Vercel polling (CSV)	`remote/vercel_poller`

Databases

Variable	Default	Description	Module
`MONGODB_CONNECTION_STRING`		MongoDB connection string	`integrations/mongodb`, `integrations/_catalog_impl`
`MONGODB_DATABASE`		MongoDB database name	`integrations/mongodb`, `integrations/_catalog_impl`
`MONGODB_AUTH_SOURCE`	`admin`	MongoDB auth source	`integrations/mongodb`, `integrations/_catalog_impl`
`MONGODB_TLS`	`true`	Enable MongoDB TLS	`integrations/mongodb`, `integrations/_catalog_impl`
`MONGODB_ATLAS_PUBLIC_KEY`		MongoDB Atlas public API key	`integrations/_catalog_impl`
`MONGODB_ATLAS_PRIVATE_KEY`		MongoDB Atlas private API key	`integrations/_catalog_impl`
`MONGODB_ATLAS_PROJECT_ID`		MongoDB Atlas project ID	`integrations/_catalog_impl`
`MONGODB_ATLAS_BASE_URL`	`https://cloud.mongodb.com/api/atlas/v2`	MongoDB Atlas API base URL	`.env.example`
`REDIS_HOST`		Redis host	`integrations/redis`, `integrations/_catalog_impl`
`REDIS_PORT`	`6379`	Redis port	`integrations/redis`, `integrations/_catalog_impl`
`REDIS_USERNAME`		Redis ACL username (Redis 6+)	`integrations/redis`, `integrations/_catalog_impl`
`REDIS_PASSWORD`		Redis password	`integrations/redis`, `integrations/_catalog_impl`
`REDIS_DATABASE`	`0`	Redis database number	`integrations/redis`, `integrations/_catalog_impl`
`REDIS_SSL`	`false`	Connect using TLS	`integrations/redis`, `integrations/_catalog_impl`
`POSTGRESQL_HOST`		PostgreSQL host	`integrations/_catalog_impl`
`POSTGRESQL_PORT`	`5432`	PostgreSQL port	`integrations/_catalog_impl`
`POSTGRESQL_DATABASE`		PostgreSQL database name	`integrations/_catalog_impl`
`POSTGRESQL_USERNAME`	`postgres`	PostgreSQL username	`integrations/_catalog_impl`
`POSTGRESQL_PASSWORD`		PostgreSQL password	`integrations/_catalog_impl`
`POSTGRESQL_SSL_MODE`	`prefer`	PostgreSQL SSL mode	`integrations/_catalog_impl`
`MYSQL_HOST`		MySQL host	`integrations/_catalog_impl`
`MYSQL_PORT`	`3306`	MySQL port	`integrations/_catalog_impl`
`MYSQL_DATABASE`		MySQL database name	`integrations/_catalog_impl`
`MYSQL_USERNAME`	`root`	MySQL username	`integrations/_catalog_impl`
`MYSQL_PASSWORD`		MySQL password	`integrations/_catalog_impl`
`MYSQL_SSL_MODE`	`preferred`	MySQL SSL mode	`integrations/_catalog_impl`
`MARIADB_HOST`		MariaDB host	`integrations/_catalog_impl`
`MARIADB_PORT`	`3306`	MariaDB port	`integrations/_catalog_impl`
`MARIADB_DATABASE`		MariaDB database name	`integrations/_catalog_impl`
`MARIADB_USERNAME`		MariaDB username	`integrations/_catalog_impl`
`MARIADB_PASSWORD`		MariaDB password	`integrations/_catalog_impl`
`MARIADB_SSL`	`true`	Enable MariaDB SSL	`integrations/_catalog_impl`

Message Queues & Other Services

Variable	Default	Description	Module
`RABBITMQ_HOST`		RabbitMQ host	`integrations/_catalog_impl`
`RABBITMQ_MANAGEMENT_PORT`	`15672`	RabbitMQ management port	`integrations/_catalog_impl`
`RABBITMQ_USERNAME`		RabbitMQ username	`integrations/_catalog_impl`
`RABBITMQ_PASSWORD`		RabbitMQ password	`integrations/_catalog_impl`
`RABBITMQ_VHOST`	`/`	RabbitMQ vhost	`integrations/_catalog_impl`
`RABBITMQ_SSL`	`false`	Enable RabbitMQ SSL	`integrations/_catalog_impl`
`RABBITMQ_VERIFY_SSL`	`true`	Verify RabbitMQ SSL certificate	`integrations/_catalog_impl`
`BETTERSTACK_QUERY_ENDPOINT`		Better Stack query endpoint	`integrations/_catalog_impl`
`BETTERSTACK_USERNAME`		Better Stack username	`integrations/_catalog_impl`
`BETTERSTACK_PASSWORD`		Better Stack password	`integrations/_catalog_impl`
`BETTERSTACK_SOURCES`		Better Stack sources	`integrations/_catalog_impl`
`KAFKA_BOOTSTRAP_SERVERS`		Kafka broker addresses	`.env.example`
`KAFKA_SECURITY_PROTOCOL`	`PLAINTEXT`	Kafka security protocol	`.env.example`
`KAFKA_SASL_MECHANISM`		Kafka SASL mechanism	`.env.example`
`KAFKA_SASL_USERNAME`		Kafka SASL username	`.env.example`
`KAFKA_SASL_PASSWORD`		Kafka SASL password	`.env.example`
`DAGSTER_ENDPOINT`		Dagster GraphQL endpoint	`integrations/_catalog_impl`
`DAGSTER_API_TOKEN`		Dagster Cloud User Token (leave empty for unauthenticated local OSS)	`integrations/_catalog_impl`

Data & Search

Variable	Default	Description	Module
`OPENSEARCH_URL`		OpenSearch instance URL	`integrations/_catalog_impl`
`OPENSEARCH_USERNAME`		OpenSearch username	`integrations/_catalog_impl`
`OPENSEARCH_PASSWORD`		OpenSearch password	`integrations/_catalog_impl`
`OPENSEARCH_INDEX_PATTERN`	`*`	OpenSearch index pattern	`integrations/_catalog_impl`
`OPENSEARCH_MAX_RESULTS`	`100`	Max OpenSearch results	`integrations/_catalog_impl`
`OPENOBSERVE_URL`		OpenObserve instance URL	`integrations/_catalog_impl`
`OPENOBSERVE_TOKEN`		OpenObserve API token	`integrations/_catalog_impl`
`OPENOBSERVE_USERNAME`		OpenObserve username	`integrations/_catalog_impl`
`OPENOBSERVE_PASSWORD`		OpenObserve password	`integrations/_catalog_impl`
`OPENOBSERVE_ORG`	`default`	OpenObserve organization	`integrations/_catalog_impl`
`OPENOBSERVE_STREAM`		OpenObserve stream	`integrations/_catalog_impl`
`OPENOBSERVE_MAX_RESULTS`	`100`	Max OpenObserve results	`integrations/_catalog_impl`
`VICTORIA_LOGS_URL`		VictoriaLogs instance URL	`integrations/_catalog_impl`
`VICTORIA_LOGS_TENANT_ID`		VictoriaLogs tenant ID	`integrations/_catalog_impl`
`CLICKHOUSE_HOST`		ClickHouse host	`.env.example`
`CLICKHOUSE_PORT`	`8123`	ClickHouse port	`.env.example`
`CLICKHOUSE_DATABASE`	`default`	ClickHouse database	`.env.example`
`CLICKHOUSE_USER`	`default`	ClickHouse username	`.env.example`
`CLICKHOUSE_PASSWORD`		ClickHouse password	`.env.example`
`CLICKHOUSE_SECURE`	`false`	Enable ClickHouse TLS	`.env.example`
`SNOWFLAKE_ACCOUNT_IDENTIFIER`		Snowflake account identifier	`integrations/_catalog_impl`
`SNOWFLAKE_ACCOUNT`		Snowflake account (alternative)	`integrations/_catalog_impl`
`SNOWFLAKE_TOKEN`		Snowflake token	`integrations/_catalog_impl`
`SNOWFLAKE_USER`		Snowflake username	`integrations/_catalog_impl`
`SNOWFLAKE_PASSWORD`		Snowflake password	`integrations/_catalog_impl`
`SNOWFLAKE_WAREHOUSE`		Snowflake warehouse	`integrations/_catalog_impl`
`SNOWFLAKE_ROLE`		Snowflake role	`integrations/_catalog_impl`
`SNOWFLAKE_DATABASE`		Snowflake database	`integrations/_catalog_impl`
`SNOWFLAKE_SCHEMA`		Snowflake schema	`integrations/_catalog_impl`
`SNOWFLAKE_MAX_RESULTS`	`50`	Max Snowflake results	`integrations/_catalog_impl`
`AZURE_SQL_SERVER`		Azure SQL server name	`integrations/_catalog_impl`
`AZURE_SQL_DATABASE`		Azure SQL database name	`integrations/_catalog_impl`
`AZURE_SQL_PORT`		Azure SQL port	`integrations/_catalog_impl`
`AZURE_SQL_USERNAME`		Azure SQL username	`integrations/_catalog_impl`
`AZURE_SQL_PASSWORD`		Azure SQL password	`integrations/_catalog_impl`
`AZURE_SQL_DRIVER`	`ODBC Driver 18 for SQL Server`	Azure SQL ODBC driver	`integrations/_catalog_impl`
`AZURE_SQL_ENCRYPT`	`true`	Enable Azure SQL encryption	`integrations/_catalog_impl`
`AZURE_LOG_ANALYTICS_WORKSPACE_ID`		Azure Log Analytics workspace ID	`integrations/_catalog_impl`
`AZURE_LOG_ANALYTICS_TOKEN`		Azure Log Analytics token	`integrations/_catalog_impl`
`AZURE_TENANT_ID`		Azure tenant ID	`integrations/_catalog_impl`
`AZURE_SUBSCRIPTION_ID`		Azure subscription ID	`integrations/_catalog_impl`
`AZURE_MAX_RESULTS`	`100`	Max Azure results	`integrations/_catalog_impl`

MCP Servers

Variable	Default	Description	Module
`OPENCLAW_MCP_URL`		OpenClaw MCP server URL	`integrations/_catalog_impl`
`OPENCLAW_MCP_COMMAND`		OpenClaw MCP command to execute	`integrations/_catalog_impl`
`OPENCLAW_MCP_MODE`	`streamable-http`	OpenClaw MCP transport mode	`integrations/_catalog_impl`
`OPENCLAW_MCP_ARGS`		OpenClaw MCP command arguments	`integrations/_catalog_impl`

Git Integrations

Variable	Description	Module
`GITHUB_EVENT_PATH`	GitHub Actions event payload path	`integrations/github/issue_comments`
`GITHUB_REPOSITORY`	GitHub repository (owner/repo)	`integrations/github/issue_comments`
`GITHUB_TOKEN`	GitHub personal access token	`.env.example`

Multi-Instance Integrations

Many integrations support multi-instance configuration via JSON:

Variable Pattern	Description
`GRAFANA_INSTANCES`	JSON array of Grafana instances
`DD_INSTANCES`	JSON array of Datadog instances
`GROUNDCOVER_INSTANCES`	JSON array of groundcover instances
`HONEYCOMB_INSTANCES`	JSON array of Honeycomb instances
`CORALOGIX_INSTANCES`	JSON array of Coralogix instances
`AWS_INSTANCES`	JSON array of AWS accounts
`ARGOCD_INSTANCES`	JSON array of Argo CD instances

See multi-instance-integrations for detailed format.

LLM Classification Models

When not specified, classification models fall back to reasoning models or provider defaults:

Variable	Default	Description
`ANTHROPIC_CLASSIFICATION_MODEL`		Anthropic classification model
`OPENAI_CLASSIFICATION_MODEL`		OpenAI classification model
`OPENROUTER_CLASSIFICATION_MODEL`		OpenRouter classification model
`DEEPSEEK_CLASSIFICATION_MODEL`		DeepSeek classification model
`GEMINI_CLASSIFICATION_MODEL`		Gemini classification model
`NVIDIA_CLASSIFICATION_MODEL`		NVIDIA classification model
`MINIMAX_CLASSIFICATION_MODEL`		MiniMax classification model
`GROQ_CLASSIFICATION_MODEL`		Groq classification model

Notes

All variables support .env file configuration
Environment variables take precedence over .env file
Variables with empty values are typically treated as “not set”
Boolean values accept: 1, true, yes, on (case-insensitive)
See LLM Providers for detailed provider setup instructions

Overview

Local

Enterprise

Environment guides

Configuration

LLM Providers

Anthropic

OpenAI

OpenRouter

DeepSeek

Google Gemini

NVIDIA

MiniMax

Groq

Azure OpenAI

Amazon Bedrock

Ollama (Local)

LLM Reasoning Effort

CLI Investigation Tools

Output & Debugging

Credentials & Authentication

When there is no system keychain

Telemetry & Monitoring

Paths & Directories

Feature Flags

Integration Credentials

Observability & Monitoring

Cloud Platforms

Messaging

Code & Collaboration

Databases

Message Queues & Other Services

Data & Search

MCP Servers

Git Integrations

Multi-Instance Integrations

LLM Classification Models

Notes

​LLM Providers

​Anthropic

​OpenAI

​OpenRouter

​DeepSeek

​Google Gemini

​NVIDIA

​MiniMax

​Groq

​Azure OpenAI

​Amazon Bedrock

​Ollama (Local)

​LLM Reasoning Effort

​CLI Investigation Tools

​Output & Debugging

​Credentials & Authentication

​When there is no system keychain

​Telemetry & Monitoring

​Paths & Directories

​Feature Flags

​Integration Credentials

​Observability & Monitoring

​Cloud Platforms

​Messaging

​Code & Collaboration

​Databases

​Message Queues & Other Services

​Data & Search

​MCP Servers

​Git Integrations

​Multi-Instance Integrations

​LLM Classification Models

​Notes

LLM Providers

Anthropic

OpenAI

OpenRouter

DeepSeek

Google Gemini

NVIDIA

MiniMax

Groq

Azure OpenAI

Amazon Bedrock

Ollama (Local)

LLM Reasoning Effort

CLI Investigation Tools

Output & Debugging

Credentials & Authentication

When there is no system keychain

Telemetry & Monitoring

Paths & Directories

Feature Flags

Integration Credentials

Observability & Monitoring

Cloud Platforms

Messaging

Code & Collaboration

Databases

Message Queues & Other Services

Data & Search

MCP Servers

Git Integrations

Multi-Instance Integrations

LLM Classification Models

Notes