Skip to content

Environment variable descriptions (raw)

7,577 distinct values, measured against 42,703 published entries.

Value Entries Share
AZURE_CLIENT_ID=Service principal client ID. Optional; required only when using ClientSecretCredential. · AZURE_CLIENT_SECRET=Service principal client secret. Optional; required only when using ClientSecretCredential. · AZURE_SUBSCRIPTION_ID=Azure subscription ID to query for cost data. Optional; auto-detected from az login session if not set. · AZURE_TENANT_ID=Azure tenant ID for service principal authentication. Optional; falls back to DefaultAzureCredential (e.g. az login) if unset. · GCP_BILLING_ACCOUNT_ID=GCP billing account ID (format: XXXXXX-XXXXXX-XXXXXX). Optional; needed only for budget monitoring. · GCP_BILLING_TABLE=Fully-qualified BigQuery billing export table (project.dataset.table_name). Required for GCP cost queries. · GCP_PROJECT_ID=Override GCP project ID if the billing dataset lives in a different project than GOOGLE_CLOUD_PROJECT. · GOOGLE_APPLICATION_CREDENTIALS=Path to GCP service account JSON key file. Optional; falls back to Application Default Credentials (gcloud auth). · GOOGLE_CLOUD_PROJECT=GCP project ID. Recommended to set explicitly in the MCP client env block — may not be inherited from gcloud shell configuration. 1 0.0 %
AZURE_DEVOPS_TOKEN=Azure DevOps PAT (Code: Read) — private projects · BITBUCKET_TOKEN=Bitbucket Cloud access token or app password — private repos, workspace queue scans · GITEA_TOKEN=Gitea/Forgejo/Codeberg token — private repos · GITHUB_TOKEN=GitHub token — private repos and higher API rate limits for vet_url/queue · GITLAB_TOKEN=GitLab token (read_api) — private projects and self-hosted instances 1 0.0 %
AZURE_IOT_HUB_AMQP_PORT=AMQP port for send_c2d_message, default 5671 · AZURE_IOT_HUB_AMQP_USE_TLS=true/false, default true · AZURE_IOT_HUB_CONNECTION_STRING=IoT Hub connection string with service/iothubowner rights, e.g. HostName=my-hub.azure-devices.net;SharedAccessKeyName=service;SharedAccessKey=... 1 0.0 %
AZURE_OPENAI_API_KEY=Azure OpenAI API key (optional). · AZURE_OPENAI_DEPLOYMENT=Azure OpenAI deployment name. · AZURE_OPENAI_ENDPOINT=Azure OpenAI endpoint hostname. · GEMINI_API_KEY=Gemini API key (alternative to GOOGLE_API_KEY). · GOOGLE_API_KEY=Google GenAI API key for Veo video tools. · GOOGLE_CLOUD_LOCATION=GCP region for Vertex AI (e.g. us-central1). · GOOGLE_CLOUD_PROJECT=GCP project id for Vertex AI. · GOOGLE_GENAI_USE_VERTEXAI=Set to true to use Vertex AI for Google GenAI. · MEDIA_GEN_DIRS=Comma-separated output directories for media files. · MEDIA_GEN_MCP_TEST_SAMPLE_DIR=Path to sample images for the test-images tool. · MEDIA_GEN_MCP_URL_PREFIXES=Public URL prefixes for mapping local output files. · MEDIA_GEN_URLS=Comma-separated URL prefixes allowed for fetch tools. · OPENAI_API_KEY=OpenAI API key (required for OpenAI image/video tools). · OPENAI_API_VERSION=Azure OpenAI API version. 1 0.0 %
AZURE_OPENAI_API_KEY=Azure OpenAI API key — enables vector search via Azure (requires AZURE_OPENAI_ENDPOINT) · AZURE_OPENAI_ENDPOINT=Azure OpenAI endpoint URL · AZURE_STORAGE_ACCOUNT=azure-blob staging via DefaultAzureCredential (managed identity) — alternative to the connection string · AZURE_STORAGE_CONNECTION_STRING=azure-blob staging: storage account connection string (account-key SAS) · AZURE_STORAGE_CONTAINER=Blob container for staged uploads · EMBEDDING_MODEL=Embedding model or Azure deployment name · ITGLUE_ADVANCED_TOOLSET=true/1 registers the advanced toolset (itglue_get, itglue_find_endpoint) — an opt-in read-only escape hatch; password resources stay blocked · ITGLUE_API_KEY=IT Glue API key (Account → Settings → API Keys) · ITGLUE_BASE_URL=Full API base URL (overrides ITGLUE_REGION) · ITGLUE_REGION=IT Glue region: us, eu, or au · OPENAI_API_KEY=OpenAI API key — enables semantic vector search tools · PUBLIC_BASE_URL=Externally reachable base URL of this server, used for memory-mode upload URLs · UPLOAD_STAGING=Staged file uploads for upload_id: off, memory (PUT to this server's /upload/:id), or azure-blob (SAS upload to Azure Blob Storage). Default: memory on http, off on stdio · VECTOR_INDEX_PATH=Path of the persisted vector index file 1 0.0 %
AZURE_OPENAI_API_KEY=Azure OpenAI API key — enables vector search via Azure (requires AZURE_OPENAI_ENDPOINT) · AZURE_OPENAI_ENDPOINT=Azure OpenAI endpoint URL · EMBEDDING_MODEL=Embedding model or Azure deployment name · ITGLUE_ADVANCED_TOOLSET=true/1 registers the advanced toolset (itglue_get, itglue_find_endpoint) — an opt-in read-only escape hatch; password resources stay blocked · ITGLUE_API_KEY=IT Glue API key (Account → Settings → API Keys) · ITGLUE_BASE_URL=Full API base URL (overrides ITGLUE_REGION) · ITGLUE_REGION=IT Glue region: us, eu, or au · OPENAI_API_KEY=OpenAI API key — enables semantic vector search tools · VECTOR_INDEX_PATH=Path of the persisted vector index file 1 0.0 %
AZURE_ORG=Azure DevOps organization name · AZURE_PAT=Personal access token (Work Items: Read & Write) · AZURE_PROJECT=Azure DevOps project name 1 0.0 %
Amtsblick__Kontakt=Kontaktadresse des Betreibers für den User-Agent (E-Mail). Ohne sie werden keine Pegeldaten abgerufen. 1 0.0 %
B2BLEADS_API_KEY=Your B2BLeads API key, from https://b2bleadsapi.com/app/integrations · B2BLEADS_API_URL=Override the B2BLeads API base URL (defaults to https://api.b2bleadsapi.com) 1 0.0 %
B2_APPLICATION_KEY=Backblaze B2 application key secret. · B2_APPLICATION_KEY_ID=Backblaze B2 application key ID for native B2 and S3-compatible tools. · B2_MASTER_KEY=Optional master key secret for Partner/Groups API tools. · B2_MASTER_KEY_ID=Optional master key ID for Partner/Groups API tools. · B2_REGION=Optional fallback S3-compatible region used before authorization. 1 0.0 %
BABELWRAP_API_KEY=Your BabelWrap API key (get one at babelwrap.com) 1 0.0 %
BABELWRAP_API_KEY=Your BabelWrap API key. Get one at https://babelwrap.com 1 0.0 %
BACKBURNER_ALLOW=Comma-separated regexes; if set, only matching commands may run · BACKBURNER_DENY=Comma-separated regexes; matching commands are always refused (deny wins over allow) · BACKBURNER_HOME=Directory for the task database and log files (default: ~/.backburner) 1 0.0 %
BACKEND=Container backend to use. Must match the installed extra: mcp-docker, mcp-podman, or mcp-k8s. · COMMIT_CONTAINER=Commit the container after a run so installed libraries persist between sessions. · DOCKER_HOST=Docker or Podman socket URL, e.g. unix:///var/run/docker.sock · KEEP_TEMPLATE=Keep the base image after the session ends to avoid re-pulling it on the next run. · KUBECONFIG=Path to kubeconfig file when BACKEND=kubernetes. · NAMESPACE=Kubernetes namespace used for sandbox pods when BACKEND=kubernetes. · SANDBOX_CAP_DROP=Comma-separated Linux capabilities to drop. Recommended: ALL. Docker and Podman backends only. · SANDBOX_CPUS=Fractional CPU allocation for the sandbox container, e.g. 1.5. Docker and Podman backends only. · SANDBOX_MEMORY=Memory limit for the sandbox container, e.g. 4g. Docker and Podman backends only. · SANDBOX_NETWORK_MODE=Network mode for the sandbox container. Set to 'none' for hardened isolation. Docker and Podman backends only. · SANDBOX_READ_ONLY=Mount the sandbox root filesystem read-only. Recommended: true. Docker and Podman backends only. · SANDBOX_SECURITY_OPT=Comma-separated container security options, e.g. no-new-privileges. Docker and Podman backends only. 1 0.0 %
BACKENDGEN_ALLOWED_ROOTS=Optional list of filesystem roots BackendGen may read and write, separated by ';' (or ':' between POSIX absolute paths). 1 0.0 %
BACKEND_API_KEY=API key or auth token. Required for Traceloop, Datadog, Sentry, New Relic, and Honeycomb; not used by Jaeger, Tempo, or X-Ray. For New Relic this must be a User API key, not an Ingest or License key; for Honeycomb this must be a Configuration Key, not an Ingest or Management key. · BACKEND_APP_KEY=Datadog Application key, required in addition to BACKEND_API_KEY for the Datadog backend only. · BACKEND_AWS_REGION=AWS region, e.g. us-east-1. Required for the X-Ray backend only - X-Ray authenticates via boto3's standard AWS credential chain, not BACKEND_API_KEY. · BACKEND_ENVIRONMENTS=Comma-separated environments for the Traceloop backend only. · BACKEND_HONEYCOMB_DATASET=Honeycomb dataset slug, required for the Honeycomb backend only - the Query Data API is dataset-scoped. · BACKEND_NEWRELIC_ACCOUNT_ID=New Relic account ID, required for the New Relic backend only - NerdGraph queries are user-scoped, not account-scoped, so this identifies which account's data to read. · BACKEND_SENTRY_ORG=Sentry organization slug, required for the Sentry backend only. · BACKEND_SENTRY_PROJECT=Sentry project slug, optional for the Sentry backend - narrows queries to one project. · BACKEND_TEMPO_INSTANCE_ID=Grafana Cloud stack/instance ID, used with BACKEND_API_KEY for Basic Auth (Tempo backend only, required for Grafana Cloud-hosted Tempo, not needed for self-hosted Tempo). · BACKEND_TIMEOUT=Request timeout in seconds. · BACKEND_TYPE=Which observability backend to query. · BACKEND_URL=Backend API endpoint, e.g. http://localhost:16686 for Jaeger, https://api.datadoghq.com for Datadog, https://api.newrelic.com/graphql for New Relic, or https://api.honeycomb.io for Honeycomb. For the X-Ray backend this is decorative only (e.g. https://xray.us-east-1.amazonaws.com) - never dereferenced for a live request. · LOG_LEVEL=Logging level. · MAX_TRACES_PER_QUERY=Maximum traces validated per query (1-1000). · MCP_INCLUDE_ARGS_IN_SPANS=Include tool call arguments/results as OTel span attributes when self-instrumentation is enabled. Off by default since they may contain sensitive data. · OTEL_EXPORTER_OTLP_ENDPOINT=Enables opt-in OTel self-instrumentation of tool calls when set. Unset means zero overhead - no TracerProvider is configured and no middleware is registered. · OTEL_SERVICE_NAME=Service name reported in self-instrumentation spans, only used when OTEL_EXPORTER_OTLP_ENDPOINT is set. · SLOW_REQUEST_THRESHOLD_MS=Logs a WARNING for any backend request slower than this, independent of LOG_LEVEL. Unset disables slow-request logging. 1 0.0 %
BACKEND_API_URL=Base URL for the Sentimatix REST API. Defaults to the hosted production instance. · SENTIMATIX_API_KEY=Optional API key for authenticated tier access. Leave blank for free tier. 1 0.0 %
BACKEND_INVESTIGATOR_TELEMETRY=Set false to force off optional usage statistics. No events before explicit consent. true does not grant consent. · BACKEND_INVESTIGATOR_WORKSPACE=Optional absolute Linux/WSL application root initialized with setup. Restricts tools and history to its bound project. Omit only for intentional shared mode. 1 0.0 %
BACKLOG_API_KEY=Your Backlog API key from account settings · BACKLOG_HOST=Your Backlog instance URL (e.g., your-space-id.backlog.com) 1 0.0 %
BACKTEST360_API_KEY=Backtest360 engine API key, sent to the engine as X-API-Key. Get a free key at https://backtest360.com/api-access. 1 0.0 %
BACKWORK_API_BASE=Override the Backwork API base URL. Defaults to https://backworkhealth.com/api/v1. · BACKWORK_API_KEY=Backwork API key (bwk_live_...). The hosted remote uses OAuth instead and needs no key. 1 0.0 %
BADGE_API_KEY=Badge API key (create under Settings -> API keys). Required for screen_agent; optional for read-only tools. · BADGE_API_URL=Badge REST API base URL. Defaults to https://api.badgeia.com. · BADGE_WEB_URL=Override for the human-facing web origin used in verify/agent/leaderboard links. Auto-derived from BADGE_API_URL when unset. 1 0.0 %
BAGS_ALLOW_UNCAPPED_TOKEN_SWAPS=Permit non-SOL-input swaps, which the SOL-denominated caps cannot limit. · BAGS_ALLOW_UNCONFIRMED=Set true to skip the two-step confirmation on writes. Default false. · BAGS_API_KEY=Bags API key from dev.bags.fm. Required for all tools. · BAGS_KEYPAIR_PATH=Path to a Solana keypair JSON file. Only required for write tools. · BAGS_MAX_SOL_PER_SESSION=Hard cap on total SOL spent across one server session. Default 1.0. · BAGS_MAX_SOL_PER_TX=Hard cap on SOL spent by any single transaction. Default 0.1. · BAGS_NETWORK=Solana cluster: devnet (default) or mainnet. Writes on mainnet move real funds. · BOS_REQUIRED_BALANCE=Minimum BOS balance required to call write tools. Default 10000. · BOS_TOKEN_MINT=Mint address of the BOS token used for gating write tools. · SOLANA_RPC_URL=Custom Solana RPC endpoint. Defaults to the public RPC for BAGS_NETWORK. · USE_MOCK_DATA=Makes bags_get_claimable_fees return fabricated data instead of calling the API. Default false. 1 0.0 %
BAILEYS_SESSION_DIR=Baileys credentials and synchronized history directory (default: <working directory>/baileys-sessions). Use a private absolute path, owned by one server process. · BROWSER_EXECUTABLE_PATH=webjs only: browser executable. Chrome and Edge are auto-detected; set this to use a specific browser. Chrome/Edge is required for sending videos and GIFs. · FFMPEG_PATH=Path to an ffmpeg binary (only needed to override the ffmpeg installed automatically via the ffmpeg-static npm package) · HEALTH_CHECK_INTERVAL_MS=webjs only: connection health check interval in milliseconds (<=0 disables) · LOG_LEVEL=Log level for the log file (logs/mcp-whatsapp.log) · LOG_STDERR_LEVEL=Log level for stderr (visible in most MCP client logs) · MCP_HTTP_HOST=Bind address for the HTTP endpoint (do not expose publicly; the endpoint controls your personal WhatsApp account) · MCP_HTTP_PORT=When set, additionally exposes the MCP endpoint at http://127.0.0.1:<port>/mcp (Streamable HTTP) from the same process · MCP_OAUTH=Enable OAuth for the HTTP endpoint: /mcp requires a bearer token and clients authenticate via a browser page showing the WhatsApp QR code / pairing form · TOOL_READY_TIMEOUT_MS=How long tool calls wait for the WhatsApp client to become ready before erroring · WA_WEB_VERSION=webjs only: pin the WhatsApp Web version if a WhatsApp-side update breaks the client. · WHATSAPP_BACKEND=WhatsApp backend: webjs uses a persistent browser profile; baileys connects without a browser and requires separate pairing · WHATSAPP_HEADLESS=webjs only: set to false to show the browser window instead of running headless · WHATSAPP_NO_SANDBOX=webjs only: explicitly disable the browser sandbox for containers that require it. The sandbox is enabled by default. · WHATSAPP_PAIRING_PHONE_NUMBER=Phone number pairing: when set (international, symbol-free format, e.g. 4915112345678), an 8-character pairing code is logged to stderr on startup instead of relying on the QR code · WHATSAPP_SESSION_DIR=webjs only: persistent browser profile directory (default: <working directory>/whatsapp-sessions) 1 0.0 %
BAILINGHUB_ALLOW_INSECURE_HTTP=Optional explicit override for operator-controlled private HTTP networks. · BAILINGHUB_BASE_URL=Absolute URL of the self-hosted BailingHub deployment. · BAILINGHUB_CLIENT_TOKEN=Dedicated BailingHub Client Token restricted to the configured route. · BAILINGHUB_ROUTE=One allowlisted BailingHub route fixed by the operator for this server process. 1 0.0 %
BALANCE_API_KEY=Account key from join.html (be_…) for your personal timetable. Leave empty for free tools (today/body/adjust/month/fold/sky/almanac/solar-terms). Empty key cannot call personal_day / personal_month. · BALANCE_ENGINE_API_BASE=Optional API origin. Default https://balance-engine.vercel.app 1 0.0 %
BALE_ALLOWED_CHATS=Comma-separated chat ids or @usernames the bot may read and message; all other chats are ignored. Recommended. · BALE_BOT_TOKEN=Bot token from @botfather in Bale (https://ble.ir/botfather). · BALE_MCP_DB=Local message history file (default ~/.bale-mcp/messages.db); ':memory:' keeps nothing on disk. · BALE_MCP_PROXY=Optional HTTP proxy URL for every request; system proxy settings are ignored. 1 0.0 %
BALLDONTLIE_API_KEY=Ball Don't Lie API key for player stats (register free at app.balldontlie.io) · EBAY_APP_ID=eBay API application ID (register free at developer.ebay.com) · EBAY_CERT_ID=eBay API certificate ID (register free at developer.ebay.com) 1 0.0 %
BALMAS_API_KEY=BALMAS API key (bk_...) - free at balmasai.com/app/team 1 0.0 %
BAMBOO_PROXY=HTTP proxy URL, if Bamboo is only reachable via a proxy. · BAMBOO_TOKEN=Bamboo personal access token. Read permission is sufficient. · BAMBOO_URL=Base URL of your Bamboo server, e.g. https://bamboo.example.com · BITBUCKET_URL=Bitbucket base URL, used to build source links in build results. · MCP_TRANSPORT=Transport mode. Must be stdio when the client runs the container. · VERBOSE=Set to true to log requests and errors to stderr. 1 0.0 %
BAMWOR_API_KEY=Optional. Bamwor API key for full access and higher rate limits. Works anonymously without it. Get one free at https://bamwor.com/en/developers/quickstart 1 0.0 %
BANANA_PEEL_API_KEY=Banana Peel API key. Agents mint one with npx -y @banana-peel/cli init --agent 1 0.0 %
BANCO_API_URL=Base URL of the Banco de Ideas API 1 0.0 %
BANDITDB_API_KEY=API key. A writer-role key is enough for get_intuition and record_outcome; admin is needed to create or delete campaigns. · BANDITDB_URL=Base URL of the BanditDB server, e.g. http://localhost:8080 or https://your-instance.api.banditdb.com. Try it against the public sandbox at https://sandbox.banditdb.com before running your own. 1 0.0 %
BANKREGPULSE_API_URL=Custom API endpoint URL (optional, defaults to the LexRegPulse production API) 1 0.0 %
BANKSYNC_API_KEY=Your BankSync API key from https://banksync.io/developers 1 0.0 %
BANK_PROVIDER=Provider selector — "mock" (default, offline fixtures) or "paystack" for live data. · PAYSTACK_SECRET_KEY=Paystack secret key (free at paystack.com) — required when BANK_PROVIDER=paystack. 1 0.0 %
BANXICO_TOKEN=Free API token for Banxico's SIE (Sistema de Informacion Economica) 1 0.0 %
BAREVALUE_API_KEY=Optional Barevalue API key. Without one, barevalue_register gives a new email address a key. 1 0.0 %
BASEALPHA_API_KEY=Optional ba_… key. Skip and run apply_for_api_key + claim_api_key (or onboard_and_launch). · BASEALPHA_PRIVATE_KEY=Dedicated agent signing wallet on Base. Required for launch/trade/claim. Never sent to BaseAlpha. · BASEALPHA_RPC_URL=Optional Base RPC override. 1 0.0 %
BASECOAT_PROJECT_ROOT=Optional host application directory for basecoat://project/context (overridden by --project-root) 1 0.0 %
BASELINE_API_KEY=API key for the Baseline API. Contact chadmcnutt@gmail.com for a key -- self-serve signup is not yet available. · BASELINE_API_URL=Base URL of the Baseline API instance to query. 1 0.0 %
BASEROW_API_URL=Base URL of your Baserow instance, e.g. https://baserow.example.com · BASEROW_PASSWORD=Baserow account password · BASEROW_TOTP_SECRET=Base32 TOTP secret of your Baserow account's 2FA (totp) setup · BASEROW_USERNAME=Baserow account email 1 0.0 %
BASESCAN_API_KEY=BaseScan API key for source code analysis (optional — basic bytecode scan works without it) 1 0.0 %
BASESCAN_API_KEY=Basescan API key for fetching verified contract source code · ETHERSCAN_API_KEY=Etherscan API key for fetching verified contract source code · RPC_URL=RPC endpoint URL (defaults to Base mainnet) · THRYX_API_KEY=Premium API key for unlimited scans 1 0.0 %
BASE_CHAIN_ID=8453 = Base mainnet (default, matches live listings). 84532 = Base Sepolia testnet. · BASE_RPC_URL=Base RPC endpoint. Defaults to Base mainnet public RPC; Alchemy/QuickNode recommended for higher rate limits. · MARKET_API_URL=Marketplace API base URL. · MCP_BUYER_PRIVATE_KEY=Funded buyer wallet private key (0x-prefixed). Required for pay_and_call; read-only tools work without it. · MCP_MAX_PRICE=Hard cap on per-call cost (USDC). Refuses to pay above this. 1 0.0 %
BASE_DIR=Base directory for document storage (defaults to current working directory) · CACHE_DIR=Directory where Transformers.js models are cached (defaults to ./models/) · DB_PATH=Path to LanceDB database directory (defaults to ./lancedb/) · MODEL_NAME=Embedding model name (defaults to Xenova/all-MiniLM-L6-v2) · RAG_GROUPING=Grouping mode: 'similar' (top group only) or 'related' (top 2 groups) · RAG_HYBRID_WEIGHT=Keyword boost strength for hybrid search. 0.0 = semantic only, 1.0 = keyword only. Default 0.6 1 0.0 %
BASE_DIR=Base directory for document storage (defaults to current working directory). Ignored when BASE_DIRS is set. · BASE_DIRS=JSON array of base directories (e.g. '["/a","/b"]'). Takes precedence over BASE_DIR. · CACHE_DIR=Directory where Transformers.js models are cached (defaults to ./models/) · CHUNK_MIN_LENGTH=Minimum chunk length in characters (1-10000, defaults to 50). Chunks shorter than this threshold are filtered out during ingestion. · DB_PATH=Path to LanceDB database directory (defaults to ./lancedb/) · EMBED_HEADING_PREFIX=Add section headings to chunk embeddings when they fit (defaults to false). Independent of EMBED_TITLE_PREFIX. Re-ingest documents after changing it. · EMBED_TITLE_PREFIX=Embed each chunk together with its document title, which can help when passages don't restate the topic the title names (defaults to false). After changing it, use a new DB_PATH or delete the index and re-ingest. · HF_ENDPOINT=Hugging Face model download endpoint. Set this to a mirror URL when direct downloads are blocked (defaults to https://huggingface.co). · MAX_FILE_SIZE=Maximum file size in bytes (defaults to 104857600 / 100MB) · MODEL_NAME=Embedding model name (defaults to Xenova/all-MiniLM-L6-v2) · RAG_DEVICE=Execution device for the embedder (defaults to cpu). Passed straight to ONNX Runtime; see the Transformers.js device source for the supported backend names. If the requested device fails to initialize, the server throws an error. · RAG_DTYPE=Embedding quantization dtype for the embedder (defaults to fp32). Opt-in and pass-through; accepts any dtype the chosen model provides (fp32, fp16, q8, int8, ...). If the model has no variant for the requested dtype, the server throws an error. Changing this changes the embedding space — re-ingest existing data. · RAG_GROUPING=Grouping mode for quality filtering. 'similar' returns only the most similar group (stops at first distance jump). 'related' includes related groups (stops at second distance jump). Unset means no grouping filter · RAG_HYBRID_WEIGHT=Keyword boost factor for hybrid search (0.0-1.0, defaults to 0.6). 0 means semantic similarity only; higher values increase the keyword-match contribution to the final score. · RAG_MAX_DISTANCE=Maximum distance threshold for filtering search results. Results with distance greater than this value will be excluded. Lower values mean stricter filtering (e.g., 0.5 for high relevance only) · RAG_MAX_FILES=Maximum number of files to keep in search results. Results are filtered to include only chunks from the top N best-scoring files. For example, 1 returns only the single best-matching file's chunks. Unset means no file filtering. · RAG_RERANK_CMD=External reranker command template. Use {query} and {top} for query text and result count; unset disables reranking. · RAG_RERANK_TIMEOUT_MS=Time budget per rerank call in milliseconds (100-600000, defaults to 10000). On timeout the spawned command is killed and the pre-rerank ordering is returned. · STORE_IMAGES=Store supported PDF and DOCX images during ingestion and return them with matched chunks (defaults to false). 1 0.0 %
BASE_DIR=Base directory for document storage (defaults to current working directory). Ignored when BASE_DIRS is set. · BASE_DIRS=JSON array of base directories (e.g. '["/a","/b"]'). Takes precedence over BASE_DIR. · CACHE_DIR=Directory where Transformers.js models are cached (defaults to ./models/) · CHUNK_MIN_LENGTH=Minimum chunk length in characters (1-10000, defaults to 50). Chunks shorter than this threshold are filtered out during ingestion. · DB_PATH=Path to LanceDB database directory (defaults to ./lancedb/) · HF_AUTO_MIRROR=Enable automatic mirror detection for HuggingFace downloads (defaults to true). Set to "false" or "0" to disable and use huggingface.co directly. The mirror chain is: huggingface.co → hf-mirror.com → modelscope.cn. · HF_ENDPOINT=Explicit HuggingFace endpoint URL to use for model downloads. When set, auto-mirror detection is skipped and this URL is used directly. Useful when you have a full-featured mirror. · HTTPS_PROXY=HTTPS proxy URL for downloading embedding models (e.g. http://proxy:8080). Required only if behind a corporate proxy. · HTTP_PROXY=HTTP proxy URL for downloading embedding models. Fallback if HTTPS_PROXY is not set. · MAX_FILE_SIZE=Maximum file size in bytes (defaults to 104857600 / 100MB) · MODEL_NAME=Embedding model name (defaults to Xenova/all-MiniLM-L6-v2) · RAG_DEVICE=Execution device for the embedder (defaults to cpu). Passed straight to ONNX Runtime; see the Transformers.js device source for the supported backend names. If the requested device fails to initialize, the server throws an error. · RAG_DTYPE=Embedding quantization dtype for the embedder (defaults to fp32). Opt-in and pass-through; accepts any dtype the chosen model provides (fp32, fp16, q8, int8, ...). If the model has no variant for the requested dtype, the server throws an error. Changing this changes the embedding space — re-ingest existing data. · RAG_GROUPING=Grouping mode for quality filtering. 'similar' returns only the most similar group (stops at first distance jump). 'related' includes related groups (stops at second distance jump). Unset means no grouping filter · RAG_HYBRID_WEIGHT=Keyword boost factor for hybrid search (0.0-1.0, defaults to 0.6). 0 means semantic similarity only; higher values increase the keyword-match contribution to the final score. · RAG_MAX_DISTANCE=Maximum distance threshold for filtering search results. Results with distance greater than this value will be excluded. Lower values mean stricter filtering (e.g., 0.5 for high relevance only) · RAG_MAX_FILES=Maximum number of files to keep in search results. Results are filtered to include only chunks from the top N best-scoring files. For example, 1 returns only the single best-matching file's chunks. Unset means no file filtering. · RAG_WATCH=Enable automatic file watcher to detect changes and re-index modified files (true/1 to enable). Useful during active development. 1 0.0 %
BASE_PRIVATE_KEY=0x-prefixed EVM private key, wallet funded with Base USDC -- signs EIP-3009 payments locally. 1 0.0 %

This sheet lists 50 of 7,577 values; they cover 50 entries.

Sheet 26 of 152, 7,577 values in total.

Counted from the projection, the same table the list filters on. An entry counts once per value.

tracevero · https://tracevero.com/werkzeuge/merkmal/mcp/roh_umgebungsbeschreibungen?seite=26