Zum Inhalt springen

Beschreibungen der Umgebungsvariablen (Rohangabe)

7.577 verschiedene Werte, gemessen an 42.703 ausgelieferten Einträgen.

Wert Einträge Anteil
ANTHROPIC_API_KEY=Anthropic Claude API key (one of the 4 provider keys is required) · CLAUDE_CODE_OAUTH_TOKEN=Claude Code OAuth token (one of the 4 provider keys is required) · GOOGLE_API_KEY=Google Gemini API key (one of the 4 provider keys is required) · OPENAI_API_KEY=OpenAI GPT API key (one of the 4 provider keys is required) 1 0.0 %
ANTHROPIC_API_KEY=Anthropic Claude API key (optional, at least one provider required) · GLM_API_KEY=Zhipu GLM API key (optional, at least one provider required) · GOOGLE_API_KEY=Google Gemini API key (optional, at least one provider required) · GROQ_API_KEY=Groq API key (optional, at least one provider required) · OPENAI_API_KEY=OpenAI API key (optional, at least one provider required) · PROMPTHEUS_MODEL=Override model selection for the chosen provider · PROMPTHEUS_PROVIDER=Override provider selection (gemini, anthropic, openai, groq, qwen, glm) · QWEN_API_KEY=Alibaba Qwen API key (optional, at least one provider required) 1 0.0 %
ANTHROPIC_API_KEY=Card-billed Anthropic API key (optional; alternative/addition to the subscription CLIs). · CLAUDE_CODE_ENABLED=Set to 1 to use your Claude Code subscription CLI (claude -p) as a provider. · CODEX_ENABLED=Set to 1 to use your Codex CLI (codex exec) as a provider. · OPENAI_COMPAT_BASE_URL=Any OpenAI-compatible endpoint (Gemini / Groq / OpenRouter / DeepSeek / Ollama). 1 0.0 %
ANTHROPIC_API_KEY=Enables Claude-powered fact extraction; falls back to the built-in regex extractor when unset. · ATTESTARI_DATABASE_URL=Postgres DSN. Set it to use Postgres + pgvector instead of the default local SQLite file (concurrent access, indexed hybrid search). · ATTESTARI_KEK=Base64 32-byte root key. Set it to enable encryption at rest and cryptographic erasure (forget() destroys the subject's key and signs the certificate). Without it, forget() is a logical delete and certificates are left unsigned. · ATTESTARI_MCP_ALLOW_FORGET=Set to 1 to offer the forget_subject tool (off by default, since erasure is irreversible and an agent can be steered by what it reads). When on, it is flagged destructive and takes two calls: a preview that returns a manifest hash, then a confirmation that passes it back. · ATTESTARI_SQLITE_PATH=Where the default SQLite ledger lives. 1 0.0 %
ANTHROPIC_API_KEY=Enables the LLM-assisted compile passes (rewrite, consolidate). Without it, toolc serves a mirror of the original tool surface. 1 0.0 %
ANTHROPIC_API_KEY=Enables the two opt-in model tools, describe_scene and narrate_change. Every other tool is deterministic and calls no model. · OPENAI_API_KEY=Alternative provider for the same two model tools, and the encoder behind find_similar_text. · UMBRA_CANOPY_TOKEN=Canopy API token. Set it and the same search tools query Umbra's authenticated commercial archive instead of the open bucket; leave it unset for the open data, which needs no credentials. · UMBRA_INDEX_DB=Path to a local catalog index (built with 'umbra index build' or fetched with 'umbra index fetch'). Without one the server walks the public bucket per search, which is correct but slow. 1 0.0 %
ANTHROPIC_API_KEY=Judge model API key. Not needed if you set RETRIEVAL_JUDGE_BACKEND=ollama to run the judge locally. · RETRIEVAL_BUDGET_USD=Spend cap for judge calls. Unset means no cap. · RETRIEVAL_JUDGE_BACKEND=Which judge to score with: anthropic (default), openai, or ollama for a fully local judge. · RETRIEVAL_JUDGE_MODEL=Judge model id, e.g. claude-sonnet-4-6 or qwen2.5:14b for Ollama. 1 0.0 %
ANTHROPIC_API_KEY=Only for the optional `looks ok` vision check. Every other check is deterministic and needs no key. 1 0.0 %
ANTHROPIC_API_KEY=Only needed for sites on no known platform, where a recipe has to be written once. Shopify and WooCommerce stores work without it. · SCRAPEWRIGHT_CACHE=Where compiled recipes are kept. Point it at durable storage, or every restart pays to work the same sites out again. Defaults to ~/.scrapewright/recipes.json · SCRAPEWRIGHT_OBEY_ROBOTS=Set to 0 to ignore robots.txt. Only reasonable on sites you own or have written permission for; the default is to obey. 1 0.0 %
ANTHROPIC_API_KEY=Only required for the 8 AI-eval tools (zpl_consistency_test, zpl_sycophancy_score, etc). Session capped at 100 Claude calls per process. · ZPL_API_KEY=User API key from zeropointlogic.io/dashboard/api-keys. Format: zpl_u_ + 48 hex chars. Service keys (zpl_s_) are rejected since v3.5.0. · ZPL_ENGINE_URL=Engine base URL. Defaults to https://engine.zeropointlogic.io. · ZPL_MODE=pure (default) hides AIN score from the AI on text-evaluation tools to prevent observer effect. coach exposes the score. 1 0.0 %
ANTHROPIC_API_KEY=Optional (Ultra AI Analyst only). BYOK Claude API key for ai_daily_briefing / ai_setup_explanation. · GOLD_MCP_LICENSE_KEY=Optional Ed25519 license token. Unlocks Pro/Premium/Ultra tools. Buy at pthaicapital.io.vn/mcp/ 1 0.0 %
ANTHROPIC_API_KEY=Optional - enables LLM-powered entity extraction · WORLD_MODEL_DB_PATH=Database location · WORLD_MODEL_EXTRACTION_MODEL=Model for entity extraction · WORLD_MODEL_REASONING_MODEL=Model for complex reasoning 1 0.0 %
ANTHROPIC_API_KEY=Optional Anthropic API key for Claude scans, prompt generation, and content-gap analysis. · GEMINI_API_KEY=Optional Google AI Studio API key for Gemini visibility scans. · OPENAI_API_KEY=Optional OpenAI API key for ChatGPT visibility scans. Configure at least one provider key to run scans. · PERPLEXITY_API_KEY=Optional Perplexity API key for Perplexity visibility scans. · SERPAPI_AI_MODE_ENABLED=Optional non-secret flag. Set true to add Google AI Mode; this adds a separate SerpAPI request per prompt. · SERPAPI_API_KEY=Optional SerpAPI key for Google AI Overviews visibility scans. · XAI_API_KEY=Optional xAI API key for grounded Grok visibility scans with web search. 1 0.0 %
ANTHROPIC_API_KEY=Optional Anthropic API key; configure at least one provider at runtime. · GEMINI_API_KEY=Optional Gemini API key; configure at least one provider at runtime. 1 0.0 %
ANTHROPIC_API_KEY=Optional fallback. BYOK for suggest_manifest_fix / generate_manifest_draft when CLARX_MCP_TOKEN is unset or hosted AI is unavailable. · CLARX_MCP_TOKEN=Org-scoped Clarx Cloud API token — unlocks hosted manifest AI (Pro: unlimited; Free: 10/mo) plus scan tools (list_repos, get_scan, list_findings, get_remediation_prompt). Create in org settings. 1 0.0 %
ANTHROPIC_API_KEY=Optional — used only by the translate tools; all validator tools run with no API key · OPENAI_API_KEY=Optional — alternative provider key for the translate tools 1 0.0 %
ANTHROPIC_API_KEY=Optional, and only for the watch app's Ask card and daily insight. The MCP tools themselves need no model key: every finding is computed in code, and the model only phrases it. · GARMIN_EMAIL=The email address for your Connect account. · GARMIN_EMAIL=The email address for your Connect account. Stored only in your local .env; `npx trainbud setup` writes it for you. · GARMIN_PASSWORD=Your Connect password. · GARMIN_PASSWORD=Your Connect password. Stored only in your local .env. Accounts with MFA enabled are not supported, because the unofficial library this drives cannot complete that flow. 1 0.0 %
ANTHROPIC_API_KEY=Optional. Adds LLM-synthesized migration steps on top of the deterministic results. The server works fully without it. 1 0.0 %
ANTHROPIC_API_KEY=Optional. Enables AI-powered classification via Haiku (more accurate). Falls back to rule-based classification if not provided. Get a key at https://console.anthropic.com 1 0.0 %
ANTHROPIC_API_KEY=Optional. Enables the Claude backend for `gt summarize`, `gt distill`, and `gt eval`. Defaults to local Ollama when unset. · GEMINI_API_KEY=Optional. Enables the Gemini backend for embedding and the LLM seam. Vertex AI ADC via `GT_GEMINI_PROJECT` is also supported. · GITHUB_TOKEN=GitHub personal access token. Optional: github-twin can also pick the token up from `gh auth token` or from its own OAuth device-flow login (`gt auth login`). · GT_PATHS__DATA_DIR=Optional. Override the on-disk data directory holding the SQLite index, embeddings, and cached GitHub responses. Defaults to `$XDG_DATA_HOME/github-twin`. 1 0.0 %
ANTHROPIC_API_KEY=Optional. Improves propose_hardware when the MCP host cannot sample. · HACKSHOP_TELEMETRY=Set to 0 to turn off anonymous usage pings. 1 0.0 %
ANTHROPIC_API_KEY=Optional. Lets Bluet judge what it reads; without it every reader runs on heuristics. 1 0.0 %
ANTHROPIC_API_KEY=Optional. Required only for the smart `research_topic` tool which uses LLM-assisted synthesis. Without it, the tool degrades gracefully to a ranked-search fallback. · DATASET_PATH=Mount path inside the container for the legal-text corpus. · DATASET_PATH=Path to a normalised corpus directory or a corpus archive (legal-texts-de-*.tar.zst). Defaults to the bundled fixture; production deployments should mount /data/corpus/latest.tar.zst. · STRICT_DATASET=Set to 'true' to refuse to start if the dataset cannot be loaded (recommended for hosted deployments). · STRICT_DATASET=Set to 'true' to refuse to start if the dataset cannot be loaded. 1 0.0 %
ANTHROPIC_API_KEY=Optional. With a key Klarion adjudicates candidates itself; without one it hands them to the calling agent to judge. 1 0.0 %
ANTHROPIC_API_KEY=Optional: enables server-side vision for classify_screen / wait_for_state; without it, classification falls back to the calling agent · KVM_PILOT_HOST=Device address (KVM/BMC) when not using a config-file profile · KVM_PILOT_MCP_ALLOW_POWER=Operator opt-in enabling power control of the managed host (per-effect gate; see also _ALLOW_HID, _ALLOW_MEDIA, _ALLOW_SSH, _ALLOW_APPLIANCE, _ALLOW_EXTERNAL_WRITE) · KVM_PILOT_MCP_DRY_RUN=Rehearsal posture: destructive calls are logged, never sent (set to 1) · KVM_PILOT_MCP_READ_ONLY=Least-privilege posture: only read-only tools are registered; effect gates force-closed. Recommended for first contact (set to 1) · KVM_PILOT_PASSWD=Device password when not using a config-file profile (KVM_PILOT_HOST / KVM_PILOT_USER accompany it) · KVM_PILOT_PROFILE=Profile name from ~/.config/kvm-pilot/config.toml naming the KVM/BMC to target · KVM_PILOT_USER=Device username when not using a config-file profile 1 0.0 %
ANTHROPIC_API_KEY=Optional: enables the LLM paths (relevance gating, skill execution). · REVENUEOS_ROOT=Path to the onboarded RevenueOS workspace (defaults to ~/.revenueos). 1 0.0 %
ANTHROPIC_API_KEY=Optional; only if a stage is configured to use Anthropic. · OPENAI_API_KEY=Optional; only if an ingest or retrieval stage is configured to use OpenAI (defaults target a local Ollama). · TRIBAL_DATABASE__URL=Postgres (pgvector) connection string. Tribal is self-hosted; you provide the database. The supported, complete deployment is docker-compose.yml. · TRIBAL_PROJECT_NAME=First-run only: the name of the project bootstrapped on first start. · TRIBAL_PROJECT_REMOTE=First-run only: the git remote that identifies the project bootstrapped on first start. · TRIBAL_PUBLIC_MCP_URL=The URL clients use to reach the server (for example http://127.0.0.1:8725/mcp). Needed for OAuth dynamic client registration once the server binds a routable interface. · TRIBAL_SERVER__BIND_ADDRESS=Address the server binds inside the container. Must be 0.0.0.0 (not the loopback default) for the published port to be reachable. · TRIBAL_SERVER__TRANSPORT=Containerised transport: http or sse. The Docker image does not serve stdio; install the binary for stdio. 1 0.0 %
ANTHROPIC_API_KEY=Pay-as-you-go Anthropic API key used to run the Vigilis agent (not a Claude.ai subscription). 1 0.0 %
ANTHROPIC_API_KEY=Provider key for codna_fix's planner — any one of ANTHROPIC_API_KEY (Anthropic), OPENAI_API_KEY (OpenAI/ChatGPT), or GEMINI_API_KEY (Gemini) works (or store one with `codna key set <provider>`). Every tool also requires the one-time free `codna login` device authorization before executing; introspection (initialize/tools/list) needs no credentials. · CODNA_API_KEY=Codna key from the one-time free `codna login` (device authorization, free community license). Required to execute any of the five tools; introspection (initialize/tools/list) needs none. · GEMINI_API_KEY=Alternative provider key (Gemini) for codna_fix's planner — see ANTHROPIC_API_KEY. · GITHUB_TOKEN=Write token, only for codna_fix with open_pr=true (opens a pull request) and codna_report_bug (files an issue). · OPENAI_API_KEY=Alternative provider key (OpenAI/ChatGPT) for codna_fix's planner — see ANTHROPIC_API_KEY. 1 0.0 %
ANTHROPIC_API_KEY=Real Anthropic key for placebo-mode arms; arms only ever see placebo keys. Optional when every arm uses passthrough auth. · OPENAI_API_KEY=Real OpenAI-compatible key for placebo-mode arms. Optional when every arm uses passthrough auth. 1 0.0 %
ANTHROPIC_API_KEY=Required for LLM-driven extraction tools (extract_methods, summarize_paper, methods_repro_review). · GITHUB_TOKEN=Optional. Raises the GitHub REST API rate limit used by reproducibility assessment. · METHODS_MCP_MODEL=Override the default Claude model (default: claude-sonnet-4-6). 1 0.0 %
ANTHROPIC_API_KEY=Required only when BOYCE_PROVIDER=anthropic and the host does not supply its own model. MCP hosts (Claude Desktop, Cursor, Claude Code, Codex, Cline, Windsurf, JetBrains) do not need this. · BOYCE_DB_URL=PostgreSQL or Redshift connection string. Optional — tools work without live DB; required for live profiling and EXPLAIN pre-flight. · BOYCE_PROVIDER=LLM provider for the NL→StructuredFilter planner (anthropic, openai). MCP hosts using their own model should not set this — the host's model does the reasoning and Boyce receives a pre-built StructuredFilter. 1 0.0 %
ANTHROPIC_API_KEY=Required only when calling list_transactions with categorize=true (DSPy transaction categorization) · CLIENT_ID=Client ID registered with the participating institution. Required unless ENVIRONMENT=mock · ENVIRONMENT='mock' (default, no credentials needed - runs entirely in-memory), 'sandbox', or 'production' · MTLS_CERT_PATH=Path to the client mTLS certificate used outside mock mode · MTLS_KEY_PATH=Path to the client mTLS private key used outside mock mode · PRIVATE_KEY_KID=Key ID matching the registered client JWKS entry outside mock mode · PRIVATE_KEY_PATH=Path to the RSA private key used for private_key_jwt and JAR signing outside mock mode · REDIS_URL=Optional Redis connection URL for persistent consent and token state 1 0.0 %
ANTHROPIC_API_KEY=Required when APITHRESHOLD_LLM_PROVIDER=anthropic. · APITHRESHOLD_LICENSE=Optional license JWT. Omit for Free tier. · APITHRESHOLD_LLM_PROVIDER=LLM provider to use: openai or anthropic. · OPENAI_API_KEY=Required when APITHRESHOLD_LLM_PROVIDER=openai. 1 0.0 %
ANTHROPIC_API_KEY=Upstream Anthropic API key. Used when the requested model is claude-* / anthropic-*. Required if you call Claude models on the free tier (BYOK). · LUCAIRN_API_KEY=Lucairn gateway API key (lcr_live_*). Issued at https://lucairn.eu/account/keys. Free tier requires BYOK upstream LLM keys (ANTHROPIC_API_KEY or OPENAI_API_KEY). · LUCAIRN_BASE_URL=Override the gateway URL. Defaults to https://gateway.lucairn.eu (production). Set this to point at a self-hosted Lucairn deployment. · LUCAIRN_TRANSPORT=Transport mode for the MCP server. One of 'direct-http' (default; v1.1+ behavior) or 'stdio-bridge' (v1.2+ opt-in). · OPENAI_API_KEY=Upstream OpenAI API key. Used when the requested model routes to OpenAI. Required if you call OpenAI models on the free tier (BYOK). 1 0.0 %
ANTHROPIC_API_KEY=Your Anthropic API key (sk-ant-...). Required for the claim extractor, red team, constitution engines, and ECS coherence checks. One v1 tool (generate_trust_card) works without an API key — it's pure assembly + deterministic audit hashing. · GLASSBOX_ECS_MODE=ECS aggregation mode: arithmetic (weighted mean, default) or geometric (stricter — any zero dimension collapses the total). · GLASSBOX_MAX_TOKENS=Per-engine-call max-tokens cap. Defaults to 2048; raise for verifying long-form content. · GLASSBOX_MODEL=Override the Claude model used by the verification engines. Defaults to claude-sonnet-4-6. 1 0.0 %
ANTHROPIC_API_KEY=Your Anthropic API key -- enables the ask_claude tool · SUPABASE_ANON_KEY=Your Supabase anon/public key · SUPABASE_URL=Your Supabase project URL (e.g. https://xxx.supabase.co) 1 0.0 %
ANTHROPIC_API_KEY=Your Anthropic API key for AI-powered testing · GHOSTQA_ALLOWED_DIRS=Colon-separated list of allowed directories for MCP server access restriction 1 0.0 %
ANTHROPIC_API_KEY=Your Anthropic API key for Claude API access (required for delegation features) 1 0.0 %
ANTHROPIC_API_KEY=Your Anthropic API key from console.anthropic.com. Required to perform translations using Claude AI. 1 0.0 %
ANTHROPIC_API_KEY=Your Anthropic API key · OLLAMA_BASE_URL=Your Ollama base URL (default: http://localhost:11434) · OPENAI_API_KEY=Your OpenAI API key 1 0.0 %
ANTHROPIC_API_KEY=Your Anthropic API key. Required — the audit engine calls Claude directly from your machine. 1 0.0 %
ANTHROPIC_API_KEY=Your own Anthropic API key. Needed only by chains that seat Anthropic models. · COHERE_API_KEY=Your own Cohere API key. Needed only by chains that seat Cohere models. · COUNCIL_MAX_USD_LIMIT=A hard per-run ceiling in US dollars that the MCP client cannot raise or remove (max_usd above it, or "none", is refused). Set it to the same value as MAX_USD_PER_RUN to lock the cap. · COUNCIL_WORKDIR=Folder for task files and run output. Default: the directory the client starts the server in. · DEEPSEEK_API_KEY=Your own DeepSeek API key. Needed only by chains that seat DeepSeek models. · GOOGLE_API_KEY=Your own Google (Gemini) API key. Needed only by chains that seat Google (Gemini) models. · GROQ_API_KEY=Your own Groq API key. Needed only by chains that seat Groq models. · MAX_USD_PER_RUN=Per-run spend ceiling in US dollars. The run stops before any stage that could pass it. Default 7. · MISTRAL_API_KEY=Your own Mistral API key. Needed only by chains that seat Mistral models. · OPENAI_API_KEY=Your own OpenAI API key. Needed only by chains that seat OpenAI models. · OPENROUTER_API_KEY=Your own OpenRouter API key. Needed only by chains that seat OpenRouter models. · TOGETHER_API_KEY=Your own Together API key. Needed only by chains that seat Together models. · ZAI_API_KEY=Your own Z.ai API key. Needed only by chains that seat Z.ai models. 1 0.0 %
ANTHROPIC_API_KEY=Your own Claude key. run_tests needs it for a test with no recording yet; a replay needs none. 1 0.0 %
ANTIDEPLOY_API_KEY=Antideploy API key for the application, created in the console under the application's settings. Starts with ad_. · ANTIDEPLOY_URL=Base URL of the Antideploy instance. Defaults to https://antideploy.com. Only set for a self-hosted or staging instance. 1 0.0 %
ANTIVIBE_MCP_PORT=Loopback port for the reader tab (default 7777) 1 0.0 %
ANUMANA_DSN=Read-only connection string for the primary target database (e.g. postgresql://readonly@host:5432/db). Optional if ANUMANA_TARGETS is used. · ANUMANA_POLICIES=JSON array of operator enforcement policies (block/warn/allow rules by risk tier or flag). Optional; open by default. · ANUMANA_TARGETS=JSON array of named targets for a multi-database setup, each with name, engine, and connection string. 1 0.0 %
ANYAPI_API_KEY=AnyAPI API key (format: aa_live_...). Mint one at https://getanyapi.com/dashboard/keys. Needed to run an API; connecting, listing the tools, and the catalog tools (list_apis, search_apis, get_api, quote_api) all work without it. 1 0.0 %
ANYPOINT_ALLOWED_ENVIRONMENTS=Comma-separated environment names or ids the server is allowed to read. Unset means all visible environments. · ANYPOINT_BASE_URL=Anypoint control plane host. Defaults to https://anypoint.mulesoft.com. · ANYPOINT_CLIENT_ID=Anypoint connected app client id (app acts on its own behalf). · ANYPOINT_CLIENT_SECRET=Anypoint connected app client secret. · ANYPOINT_MONITORING_BASE_URL=Anypoint Monitoring host. Defaults to https://monitoring.anypoint.mulesoft.com. · ANYPOINT_ORG_ID=Business group id to operate on. Defaults to the organization the credentials belong to. 1 0.0 %
ANYTYPE_API_BASE_URL=Where the Anytype desktop API is reachable from inside the container. · ANYTYPE_API_HOST_HEADER=Override the Host header sent to the Anytype API. Auto-derived when unset; Anytype rejects non-loopback hosts with 403. · ANYTYPE_API_KEY=Anytype API key, created in the desktop app under Settings -> API Keys. · ANYTYPE_MCP_SLIM_RESPONSES=Compact API responses before returning them to the client. Set to false for raw payloads. · ANYTYPE_VERSION=Anytype API version header sent with every request. · MCP_SHARED_TOKEN=Optional bearer token required on the /mcp endpoint. Empty means open, for localhost only. · PORT=Port the MCP endpoint listens on, inside the container and on the host. 1 0.0 %

Auf diesem Blatt stehen 50 von 7.577 Werten; sie decken 50 Einträge ab.

Blatt 17 von 152, 7.577 Werte insgesamt.

Gezählt wird die Projektion, dieselbe Tabelle, aus der die Liste filtert. Ein Eintrag zählt je Wert einmal.

tracevero · https://tracevero.de/werkzeuge/merkmal/mcp/roh_umgebungsbeschreibungen?seite=17