Create a new intent in a Cognigy.AI flow for NLU recognition. Validate changes with dryRun first, then train the NLU model to enable matching.
MIT
679,087 tools. Updated 2026-10-19 23:15
"How to implement intent recognition in MCP" matching MCP tools:
- Score a test utterance against trained NLU intents to evaluate recognition accuracy before deployment. Returns ranked matches with confidence scores.MIT
- Plan, implement, and review code changes in one call. Delegates tasks to CLI agents to automate the full development cycle.MIT
- Import a signed agent mandate and bind it to a Harbor intent with a replay-safe recognition proof. Authorizes delegated agent spend through the gateway.Apache 2.0
- Lists NLU connectors to view integrations with external NLU services for intent recognition, including Dialogflow, LUIS, and custom solutions.MIT
- Import a signed AgentMandateV1 through gateway v2 and bind it to a Harbor intent using a replay-safe proof.Apache 2.0
Matching MCP Servers
- AlicenseAqualityBmaintenanceEnables AI agents to search for products and prepare carts under a self-hosted, developer-controlled policy, with every purchase requiring a passkey-signed human approval before payment on the merchant's checkout.3Apache 2.0
- AlicenseNot gradedqualityFmaintenanceMCP server for intelligent circuit diagram recognition, enabling detection of components, text, topology, and netlist from circuit images.5MIT
Matching MCP Connectors
AudD MCP is a way for agents to add music recognition to apps or to fulfill song identification requests. It connects you to AudD, the music recognition API
Token intent checks; no simulation/signing. Free trial ends 2026-09-25. No NFT required.
- Export workspace context as CLAUDE.md, cursorrules, or intent-md files. Choose format and optionally specify an intent or product ID.MIT
- Preview SQL queries to assess safety, intent, and cost before execution, with automatic LIMIT applied.AGPL 3.0
- Check if the Brickognize image recognition API is online and responsive before attempting LEGO identification tasks.MIT
- Create a structured intent spec by providing a title, objective, and product ID. Use with list_intents to avoid duplicates.MIT
- Listen for the other party's next turn in a live phone conversation. Speech recognition is billed.MIT
- Get search intent classification for up to 1,000 keywords with probability scores, identifying informational, navigational, commercial, or transactional intent.Apache 2.0
- Poll a credential handoff intent to check if the human has completed it in the browser. Returns the current intent state.Apache 2.0
- Find all contracts that implement a given function signature. Identify overrides, interface implementations, and polymorphic patterns in Solidity codebases.AGPL 3.0
- Check your daily recognition credit balance to track available SPINES credits before searching books or managing reading lists.MIT
- Generate ranked solution approaches backed by documented patterns when you are unsure how to proceed. Get alternatives with rationale and next steps.MIT
- Retrieve a paginated list of async jobs for invoice generation, revenue recognition, and other background processes.MIT
- View all recognition terms configured in the TypeWhisper dictionary to review or manage custom entries.GPL 3.0
- Retrieve a numeric Sorsa Score to estimate the recognition and influence of an X/Twitter account. Provide the handle, user ID, or profile URL.MIT
- Save a structured intent spec to intent.md after building it through conversation. Define objectives, outcomes, constraints, and edge cases.MIT
- Modify an unconfirmed payment intent to adjust amount, currency, metadata, or payment settings before confirmation in OrcaRail crypto payments.MIT