"An open-source vector database for similarity search and AI applications" matching MCP tools:

• search_regulations

  Velvoite — EU Financial Regulatory Compliance

  Search the regulatory corpus using keyword / trigram matching. Uses PostgreSQL trigram similarity on document titles and summaries. Returns documents ranked by relevance with summaries and classification tags. Prefer list_documents with filters (regulation, entity_type, source) first. Only use this for free-text keyword search when structured filters aren't sufficient. Args: query: Search terms (e.g. 'strong customer authentication', 'ICT risk', 'AML reporting'). per_page: Number of results (default 20, max 100).

  Connector

• search_ens_names

  Name Whisper — ENS Intelligence Layer

  Search ENS names using natural language. Supports all query types: - Filtered search: "4-letter words under 0.1 ETH" - Concept search: "ocean themed names" (semantic similarity across 3.5M indexed ENS names) - Creative search: "names for a coffee brand" (AI-generated suggestions) - Collection search: "crypto terms expiring soon" - Activity: "what sold recently?" - Availability check: "is coffee.eth taken?" - Bulk check: "check apple.eth, banana.eth, cherry.eth" - Collection/club floor: "999 club floor", "cheapest 10k club names" (returns real listings sorted by price) Returns structured results with name, price, owner, tags, and availability info. It searches the NAME database by pattern/length/price/club/vibe — it does NOT know who real-world people, teams, brands, athletes, musicians, or films are. For "find me NBA players / pop stars / Pixar films / presidents" use enumerate_entities instead (it returns correctly-spelled labels). Use this for "floor of <club>" / "cheapest in <collection>" (find_alpha can't — it has no collection param). For lifecycle-window lists — "which names are in premium / Dutch auction", "names in grace period", "expiring soon" — use get_expiring_names instead: its grace/premium statuses are on-chain-validated and premium rows carry live pricing.

  Connector

• semantic_search

  Zoning Signal

  Semantic search across the full corpus — every place dossier, corridor signal, meeting reading, and named-pattern brief. Returns results ranked by cosine similarity in a 1024-dimensional embedding space (Voyage AI 4 + Supabase pgvector). Use when the agent does not know the canonical entity slug or named-pattern title in advance — the search returns the readings whose semantic structure best matches the natural-language query, with type, title, similarity, and resolved URL per hit. Threshold 0.55, top 12.

  Connector

• ping

  Nordic Financial MCP

  Connectivity check that confirms the Nordic MCP server process is responding. Use this at the start of a session to verify the server is reachable before making other calls. Do not use as a proxy for database health — the server can respond while the Qdrant vector database is temporarily unavailable. To confirm data availability, call search_filings directly. Returns: A greeting string: "Hello {name}! Nordic MCP server is running."

  Connector

• semantic_search

  canton-ccpedia

  PRIMARY tool for open-ended questions: how / why / what-is, troubleshooting a symptom ("why is my balance zero", "how do I fix X"), and locating config or setup steps. Conceptual/meaning-based search over the full Canton corpus (CIPs, docs, forum, mailing lists, proposals, blog, releases, ecosystem, foundation KB, YouTube) using vector+FTS hybrid retrieval with reranking. Canton-specific. Use this FIRST for anything a specific tool does not clearly own; the narrow curated tools (get_faq, find_known_issues, diagnose_error) cover only small hand-picked sets or need a literal error string, so prefer semantic_search for real how/why/config questions. Then call get_doc with a returned id to read the full source page.

  Connector

• find_similar_creatives

  CreativeScope — Mobile Game Ad Creative Intelligence

  Find visually similar creatives using the stored vector of an existing creative. For a concept without an ID, query selects an explainable seed from available creative metadata and then uses the same vector-neighbor search. For an English concept, send the original English terms only. The service resolves Chinese source-label equivalents internally before selecting the seed. Returns creative records ordered from most to least visually similar; low-similarity and near-duplicate results are excluded, and raw similarity scores are not exposed. If request_echo.seed_basis identifies a proxy seed, clearly disclose that limitation instead of presenting the results as an exact concept match. Example: 'Show variants of the toilet run viral creative concept.'

  Connector

• search_legislation

  bill-commons

  Search FULL BILL TEXT -- not just known-bill-number lookup. `q` is matched against titles, descriptions, AND ingested document text via Postgres websearch_to_tsquery (supports "quoted phrases", OR, and -exclusion, same syntax as a search engine), with a fuzzy pg_trgm title-similarity fallback when the exact query has no hits. `q` can ALSO be a bill number ("HB 123", "H.B. 123", "hb123" all match) and that fast path is tried first. Optionally filter by jurisdiction (two-letter state code or name), chamber, and status. For a curated cross-state slice of a subject (e.g. "every AI bill in the country") rather than an ad-hoc keyword search, call list_topics first -- its membership rules also match on structured subject tags this full-text search does not see.

  Connector

• verify_claim

  Ground Truth - Fresh Evidence for AI Agents

  Verify a factual claim against specific public evidence URLs before an agent repeats it or acts on it. The tool checks whether supplied keywords appear in each fetched source and reports per-source evidence. It does not perform open-ended search, semantic contradiction analysis, or prove a claim true when a page is silent; choose direct evidence URLs and interpret the result as a support signal. Fetched pages are cached for 5 minutes.

  Connector

• search_docs

  pg-aiguide

  Search documentation with hybrid semantic (vector) and keyword (BM25) search. Use semanticWeight to choose keyword-only (0), semantic-only (1), or a blend; mid values fuse rankings with RRF. Supports Tiger Cloud (TimescaleDB), PostgreSQL, and PostGIS.

  Connector

• search_servers

  MCP Marketplace

  Search the MCP Marketplace catalog. With a free-text `query` and default `sort`, results are ranked by semantic similarity (gte-small embeddings + cosine similarity), so natural-language queries like 'manage my calendar', 'something to read PDFs', or 'database for my agent' work as well as keyword searches. Each result includes `security_score` (0-10), `risk_level` (low/moderate/high/critical), `critical_findings` (count of severity=critical|high findings), pricing, rating, install count, and a URL. `ranking_mode` in the response indicates whether semantic or keyword matching was used. Before recommending an install, call get_server for full details including every flagged finding — critical_findings > 0 means the server has known security issues you must surface to the user.

  Connector

• content_similar

  gapup-mcp

  Find content entities similar to a given one. For embedded franchises this uses SEMANTIC vector similarity (pgvector) over the enrichment profile — surfacing entities that feel alike even when their tags differ literally. Falls back to shared enrichment-tag overlap for works or non-embedded entities. Each result carries a similarity score and its entity-level freshness/confidence (verifiable, sourced). When to use this tool: an agent wants recommendations or lookalikes for a franchise or work. Input: an entity_id and its type.

  Connector

• memory_recall

  SqueezeOS — Institutional AI Market Intelligence

  Memory Graph. Cost: 0.01 USDC on Base via x402 v2. Literal keyword/tag search over an agent's own stored memories, ranked by recency. Not semantic or vector search — no embedding model is involved, and the response says so explicitly (match_method field). Send PAYMENT-SIGNATURE on the MCP HTTP request; agent_wallet is optional attribution.

  Connector

• search_semantic

  Tripitaka MCP (hosted)

  Semantic search — match by meaning, not exact words. Uses vector similarity (cosine distance) over `text_pali` embedded with a multilingual MiniLM model. 🤔 **In most cases you should use `search_hybrid` instead** — it combines this semantic search with keyword search and ranks better. Use this tool only when you need: - Pure semantic results (no keyword influence) - Fine-grained `threshold` tuning (hybrid uses RRF which is harder to tune) - To debug what semantic alone picks up vs keyword ⚠️ Known limitations: - The index is **Pāli only** (English/Thai queries pass through the multilingual embedding but the model isn't tuned on Pāli) - English queries usually embed better than Thai (model is EN-primary) - For specific Pāli terms (`appamāda`, `dukkha`), exact match is better — use `search_by_keyword` instead - Pāli stock phrases recur in many suttas → similarity scores cluster; read the top 10, don't trust rank 1 alone

  Connector

• search_grants

  GrantSonar

  Search open grant opportunities in the GrantSonar corpus (federal, state, and foundation grants). Uses semantic similarity over a plain-English description of the project or need, with a keyword fallback. Returns title, agency, deadline, award amounts, similarity score, and a GrantSonar URL per hit.

  Connector

• youtuber_add

  mcp

  Add a single YouTuber to the database and trigger the full enrichment pipeline (tags, niches, description, similarity, sponsor detection). Returns a batchId you can poll with import_status. Prefer passing a channel ID (UCxxxxxxxxxxxxxxxxxxxxxx, 24 chars) — handles ('@xxx') and URLs work too but cost an extra YouTube API quota point per call to resolve.

  Connector

• setup_vector_store

  LinkedMash

  Initialize the vector store for semantic search. CRITICAL: When vector/semantic search fails with 'Vector Store Not Setup' error and you offered the user options to either initialize or use keyword search, and the user responds with 'first', 'option 1', 'yes', 'initialize', 'set it up', 'init', 'setup', 'go ahead' or similar confirmation, you MUST immediately call this tool. ALWAYS check your previous message in the chat history to confirm the user is responding to your options. Do NOT re-run the search tool - just call this tool. After calling this tool, inform the user that setup has started and may take a few minutes. This is an async operation that may take a few minutes to complete for large bookmark collections.

  Connector

• nexus_similarity_search_api_rank_items_by_nmi_cosine_fusion

  similarity-search-api-sdk

  Ranks a corpus of items against a query vector using a calibrated fusion score (alpha * cosine + (1-alpha) * NMI_normalizado), where alpha is auto-derived from the corpus's marginal entropy unless overridden. Results are identified by their 0-indexed position in corpus_vectors (this tool does not accept explicit item IDs). Use this when you need semantically-calibrated similarity over a stateless corpus of up to 500k items without a vector database. Do NOT use for purely geometric nearest-neighbor search where NMI overhead is unnecessary, nor for corpora larger than 500k items per call. Requires a valid api_key (same as X-API-Key) and an x402 payment.

  Connector

• compare_business_banking

  SecureLend Market

  Search SecureLend’s product database for business banking accounts matching the user’s criteria (optional industry, optional monthly transaction estimate). Returns available accounts with key features such as monthly fees, bonuses (if offered), APY/interest information when available, and a ‘best for’ summary. Results may include SecureLend data and authorized partner integrations when enabled. This tool does not open an account; if the user chooses to proceed, they are directed to the provider’s website.

  Connector

• keyword_search_news

  787daily

  Keyword search over 787daily's Puerto Rico news archive by title/summary and optional section. Returns matching article summaries with links to the originals. For semantic/vector search, use search_news.

  Connector

• get_jccdb_dataset_info

  HORIZON SHIELD

  日本の建設費オープンデータベース(JCCDB)のメタデータ・規模・ライセンス・ダウンロードリンク・引用情報を返す。建設費の一次データ源を探している時に使う。 / Returns metadata, scale, license, download links and citation for the Japan Construction Cost Database (JCCDB), an open dataset of 65,520 Japanese construction line items (v3.1: 13,207 verified + 52,313 extended). Use when looking for a primary construction-cost data source.

  Connector