Send a message to any of 30+ AI models (OpenAI, Anthropic, Google, Groq, xAI). Returns the model's response. Supports conversation history via the messages array.
Connector
135,121 tools. Last updated 2026-05-25 22:07
"Interacting with AI models like OpenAI or Google models" matching MCP tools:
- Run a live A/B test against the engine's TOP 3 PICKS for a stated purpose — the engine chooses the candidates from the full catalog. Generates 5 representative test queries (auto-expands to 10 or 15 if results are too close to call), runs them through the picked models in parallel, and returns real cost, latency, and plain-English commentary on who won what. Use AFTER `pick` or `rank` when the user wants the engine's own picks stress-tested with live data. DO NOT use this when the user has already named specific candidate models — the engine will ignore the names and test its own picks. Use `compare` instead in that case. Costs more than `rank` (15+ live LLM calls).Connector
- Run a live A/B test between 2–5 user-specified models for a stated purpose. NO ranking step — the supplied model_ids ARE the candidate set. Generates 5 representative test queries from the purpose, runs them through every named model in parallel, and returns real cost, latency, and plain-English commentary on who won what. Unknown IDs are dropped with a note; if fewer than 2 IDs resolve, the call refuses. Use this whenever the user names specific models to compare (e.g. 'A/B test X and Y'). For engine-chosen candidates, use `benchmark` instead. Costs more than `rank` (10+ live LLM calls). Free-tier note: when any candidate ends in ':free', the probe is capped at 3 queries (no adaptive expansion) because free-tier rate limits often push longer probes past the deploy's 5-minute ceiling — evidence will be shallower. The commentary surfaces this when it happens.Connector
- Generate an AI video and place it directly on a user's Avocado AI storyboard. Drops a 'Generating...' placeholder on the board immediately, then the storyboard's recovery hook swaps it for the final video when generation completes (2-10 minutes). Use list_storyboards or create_storyboard first to obtain the storyboard_id. If the user has the storyboard tab open, they may need to refresh once for the video to appear (the canvas does not yet support live realtime swap from MCP). Eight models supported: seedance-2.0-t2v / -t2v-fast (text only), seedance-2.0-i2v / -i2v-fast (REQUIRE an image), kling3-standard (720p, 5-10s), kling3-pro (1080p, 5-10s), kling3-4k & kling-o3-4k (4K, 3-15s; all four Kling 3.x variants support BOTH text-to-video and image-to-video). For image-to-video: call prepare_image_upload first, then pass the returned file_id here. Pricing is per-second, varies by model and resolution.Connector
- Run a live A/B test against the engine's TOP 3 PICKS for a stated purpose — the engine chooses the candidates from the full catalog. Generates 5 representative test queries (auto-expands to 10 or 15 if results are too close to call), runs them through the picked models in parallel, and returns real cost, latency, and plain-English commentary on who won what. Use AFTER `pick` or `rank` when the user wants the engine's own picks stress-tested with live data. DO NOT use this when the user has already named specific candidate models — the engine will ignore the names and test its own picks. Use `compare` instead in that case. Costs more than `rank` (15+ live LLM calls).Connector
- Run a live A/B test between 2–5 user-specified models for a stated purpose. NO ranking step — the supplied model_ids ARE the candidate set. Generates 5 representative test queries from the purpose, runs them through every named model in parallel, and returns real cost, latency, and plain-English commentary on who won what. Unknown IDs are dropped with a note; if fewer than 2 IDs resolve, the call refuses. Use this whenever the user names specific models to compare (e.g. 'A/B test X and Y'). For engine-chosen candidates, use `benchmark` instead. Costs more than `rank` (10+ live LLM calls). Free-tier note: when any candidate ends in ':free', the probe is capped at 3 queries (no adaptive expansion) because free-tier rate limits often push longer probes past the deploy's 5-minute ceiling — evidence will be shallower. The commentary surfaces this when it happens.Connector
Matching MCP Servers
- Alicense-qualityCmaintenanceProvides comprehensive AI model metadata through MCP, enabling search and filtering of 100+ AI models by capabilities, pricing, context length, and provider specifications.Last updated
- AlicenseBqualityFmaintenanceA lightweight bridge that wraps OpenAI's built-in tools (like web search and code interpreter) as Model Context Protocol servers, enabling their use with Claude and other MCP-compatible models.Last updated412MIT
Matching MCP Connectors
Focused MCP server for OpenAI image/audio generation (v2.0.0). Wraps endpoints via HAPI CLI.
Find local businesses on Google: name, address, phone, hours, ratings, and photos.
- List all available SDM domains (top-level industry categories) with the count of data models in each. Use this as the entry point when the user wants an overview of what sectors are covered, or before calling list_models_by_domain. No parameters required. Example: list_domains({})Connector
- List text-embedding models currently loaded on this node (Qwen3-Embedding, EmbeddingGemma, BGE-M3, etc.). Use list_text_embedding_catalog to browse the curated catalog.Connector
- List ASR (speech-to-text) models currently loaded on this node. Note: the audio transcriber runtime is scaffolding only — `transcribe` returns ProviderNotAvailable until the ORT-backed Whisper / Moonshine / Parakeet / Canary implementations land in the next wave.Connector
- Reverse-lookup a single concept ID (MITRE ATLAS technique like 'AML.T0051', OWASP LLM Top 10 risk like 'LLM01', OWASP Agentic Top 10 issue like 'ASI03', or ISO 42001 Annex A clause like 'A.6') across the AI Defense Matrix. Returns which framework the concept belongs to, the asset rows whose alignment cites it, the cells whose evaluation cellPrompts cite it, and those prompts themselves. Useful when a vendor's product is defined by a specific technique ('we defend AML.T0051') and they need to find which matrix cells to claim. Recognizes only concepts with structured IDs; for prose-only frameworks (NIST IR 8596, CSA AICM, Google SAIF, OWASP AI Exchange) use aidefense_get_framework_alignment instead. This server never requests your program docs or product roadmap and instructs your AI to keep them local—the matrix, framework alignments, and playbooks flow to your AI for local analysis.Connector
- Modify an existing image. REQUIRED input: exactly one of file_id OR image_url. base64 is NOT accepted — do not try to pass image bytes as a tool argument, the call will be rejected. For chat-attached images you MUST first call prepare_image_upload to get a signed PUT URL, upload the bytes there (via the inline widget on Claude.ai, or via curl on Claude Desktop / Claude Code), then call this tool with the returned file_id. For URLs the user has pasted, use image_url directly. Returns a jobId immediately; call check_job with the jobId to retrieve the edited image inline. Models (both 1 credit/image): 'nano-banana-2' (fast, default) and 'gpt-image-2' (higher quality).Connector
- List top sending sources (ESPs, ISPs, mail services) for a domain, grouped by source type. Filters: "known" (legitimate ESPs like Google, Mailgun), "unknown" (unrecognized senders), "forward" (forwarding services). Empty = all types. Returns top 20 per type with message volume, SPF/DKIM/DMARC pass/fail counts. Use this to investigate WHERE email is being sent from — especially when unknown sources appear or compliance is low. To drill down into a specific source (by IP, ISP, hostname, or reporter), use get_domain_source_details.Connector
- Enumerate ENS-friendly labels for a finite real-world entity category and report which are available vs registered. USE THIS for any query like "find me NBA hall of famers", "available Pixar films", "F1 drivers I can register", "Beatles songs that are open". The tool generates verified, correctly-spelled ENS labels — do NOT enumerate entity names from your own context and pass them to check_availability, because models routinely misspell long-tail names (scottiepippin instead of scottiepippen) or invent people who don't exist (e.g. "johncarlton" as an NBA HOFer). This tool exists precisely to avoid that. DO NOT use this for: - Vibes / themes ("luxury watch names", "edgy crypto names") — use search_ens_names with concept_search instead. - ENS-native categories ("10k club", "3-letter words") — use search_ens_names with collection_search. - Single-name lookups — use check_availability. Returns a list of entries grouped by status. Each entry has the proper name (e.g. "Scottie Pippen") alongside the ENS label (scottiepippen.eth), so you can show users the human-readable name in your reply.Connector
- Discover what's currently available in FINN's fleet. Returns all brands (with nested models), car types, fuel types, colors, subscription terms, gearshifts, and price/power/range bounds. Use this to answer questions like 'What brands does FINN offer?' or to validate filter values before searching.Connector
- List segmentation models currently loaded on this node (SAM 2, EdgeSAM, MobileSAM). Use list_segmentation_catalog to browse the curated catalog.Connector
- What's the market doing right now? Price, funding rate, CVD, whale activity, and liquidation pressure in one call — 16 fields, no LLM overhead. Feed directly into your own models or decision logic. orderflow coverage disclosed per token. REST equivalent: POST /data (0.20 USDC). Args: token: Token symbol (BTC, ETH, SOL, XRP, ADA, DOGE, AVAX, LINK, BNB, ATOM, DOT, ARB, SUI, OP, LTC, NEAR, TRX, BCH, SHIB, HBAR, TON, XLM, UNI, AAVE, AMP, ZEC)Connector
- Given a passage of text (essay, note, message, snippet, transcript), returns ~5 humans whose intellectual fingerprint matches it — recurring themes, mental models, archetypal stance, blind spots. Use when the principal asks for sparring partners, intellectual peers, "who else is wrestling with this," "who thinks like X," or "find me writers similar to this passage." Each result returns a name, three-word archetype, one-line summary, dominant themes, and a profile URL the principal can visit. The match runs over Voyage 3.5-lite text embeddings reranked by a proprietary 12-dimensional cognitive-style vector — so results align by *how* a mind reasons, not just topical overlap.Connector
- Discover available AI models with numeric IDs, tier labels, capabilities, and per-call pricing in sats. Call this before create_payment to find the right modelId for your task. Returns JSON array: [{ id, name, tier, description, price, isDefault, category }]. Models marked isDefault=true are used when you omit modelId from create_payment. Filter by category to narrow results to a specific tool. This tool is free, requires no payment, and is idempotent — safe to call repeatedly.Connector
- List all custom evaluation models for the authenticated user. Returns an array of model objects with id, name, description, and status. Use model id in artifact, rubric, and evaluation tools. Free.Connector
- List available AI models grouped by thinking level (low/medium/high). Shows default models, credit costs, capabilities for each tier. Use this before consult to understand model options.Connector