Skip to main content
Glama
650,522 tools. Updated 2026-10-14 01:26

"How to allow an LLM to ask questions to a user" matching MCP tools:

  • Ask natural-language questions about the ledger and receive free-form answers. Maps plain queries to structured filters for read-only reporting.
    MIT
  • Guides you through adding an API integration: intake your goal, ask discovery questions, route to the right workflow, and validate the contract via FetchSandbox.
    MIT
  • Evaluate an action against your hosted policies to receive a decision (allow, deny, hold, escalate) and a permit token when allowed.
    Apache 2.0
  • Check whether a consequential action may proceed before any hard-to-undo write, returning ALLOW, DENY, or STEP_UP plus an audit receipt.
    MIT
  • Ask visual questions about one or more videos, using start/end times to focus analysis on specific segments for precise answers.
    Apache 2.0
  • Evaluate a risk assessment against a policy table to decide whether to allow, ask for confirmation, or block an action.
    MIT

Matching MCP Servers

  • F
    license
    Not graded
    quality
    D
    maintenance
    Enables AI models to ask users questions through a local web interface, supporting batch questions, multi-select, and free text for human-in-the-loop interactions.
    -
  • A
    license
    Not graded
    quality
    D
    maintenance
    A minimal MCP server with get_weather and create_ticket tools, used for testing MCP servers across protocol, unit, eval, transport, and auth layers.
    MIT

Matching MCP Connectors

  • Remote MCP server for converting PDFs and scanned documents to Markdown with OCR, and Markdown to PDF. Provides asynchronous conversion jobs, result retrieval, and usage checks; requires a PDF to MD API key.

  • Extract structured JSON from any document: invoices, receipts, contracts, purchase orders, bills. Auto-detects type and returns vendor, customer, line items, totals.

  • Returns a list of PromptEye help articles to answer questions about features, settings, scores, and troubleshooting. Call this first when users ask how something works.
    MIT
  • Send a message to a company on a job application for follow-ups or questions. The user writes the message to keep it personal and specific.
    MIT
  • Send or resend a confirmation code to an unconfirmed contact to verify their address. The code is delivered as a real message, so ask the user for it to complete the verification.
    MIT
  • Answer complex relationship questions about knowledge graph data using AI interpretation. Ask natural language queries to understand connections between entities like team skills, project technologies, or personal associations.
    MIT
  • List active agent model access rules to control which LLM models each agent can use. Rules use glob patterns with deny taking precedence over allow.
    MIT
  • Retrieve the full verbatim transcript of an LLM response for a specific monitoring result. See exactly what an LLM said about your brand, including model name, prompt, and metadata.
    MIT
  • Get comparable companies and valuation multiples for a US stock to answer relative-valuation questions like 'how does this company compare to peers?'
    MIT
  • Generate a markdown interview prep sheet for a recruiter call: stage, next step, resume-to-requirement matches, gaps, sent materials, posting, expected questions, and questions to ask.
    MIT
  • Ask natural language questions about a specific hotel and get AI-generated answers based on its data, with optional web search for extra details.
    MIT