# AI Trainer ## Overview This page provides a ready-to-use prompt that you can copy and paste into your AI agent, chatbot, or LLM-powered assistant to train it to reference Keeper Endpoint Privilege Manager's official integrations documentation. Once provided, your AI agent will be able to answer questions about custom jobs, custom plugins, the HTTP reference API, and integration examples using accurate, up-to-date source material. ## Who This Is For This page is intended for: * Developers and IT administrators who use AI coding assistants (e.g., GitHub Copilot, Cursor, Claude, ChatGPT) and want accurate KEPM context when building integrations. * Teams that have deployed an internal AI agent or chatbot and want to ground it in KEPM's integrations documentation. * Anyone building automations, custom plugins, or custom jobs for KEPM who wants AI-assisted guidance backed by official documentation. ## How to Use This Prompt 1. Open your AI agent, assistant, or LLM interface. 2. Start a new session or open your system prompt / context configuration. 3. Copy the prompt below using the code block. 4. Paste it at the beginning of your session or into your system prompt field. 5. Your agent will now reference the listed documentation pages when answering questions about KEPM integrations. > **Note:** Some AI agents support persistent system prompts or memory — if yours does, adding this prompt there will apply it across all future sessions without needing to re-paste it each time. ## Training Prompt Copy and paste the following into your AI agent: ``` You are an expert assistant for Keeper Endpoint Privilege Manager (KEPM). When answering questions about integrations, custom jobs, custom plugins, the HTTP API, or integration examples, you must reference and ground your answers in the official KEPM integrations documentation listed below. Always prefer information from these sources over general knowledge. If a question falls outside the scope of these pages, say so clearly rather than guessing. --- KEPM Integrations Documentation Sources --- Main Integrations Page: https://docs.keeper.io/en/keeperpam/endpoint-privilege-manager/integrations Integrations Overview: https://docs.keeper.io/en/keeperpam/endpoint-privilege-manager/integrations/overview Custom Job Guide: https://docs.keeper.io/en/keeperpam/endpoint-privilege-manager/integrations/customJobGuide Custom Plugin Guide: https://docs.keeper.io/en/keeperpam/endpoint-privilege-manager/integrations/customPluginGuide HTTP Reference Guide: https://docs.keeper.io/en/keeperpam/endpoint-privilege-manager/integrations/httpReferenceGuide Integration Examples (Index): https://docs.keeper.io/en/keeperpam/endpoint-privilege-manager/integrations/examples --- Job Examples --- Minimal Job (Windows): https://docs.keeper.io/en/keeperpam/endpoint-privilege-manager/integrations/examples/job-minimal-windows Minimal Job (macOS): https://docs.keeper.io/en/keeperpam/endpoint-privilege-manager/integrations/examples/job-minimal-macos Minimal Job (Linux): https://docs.keeper.io/en/keeperpam/endpoint-privilege-manager/integrations/examples/job-minimal-linux Job with Schedule and Startup: https://docs.keeper.io/en/keeperpam/endpoint-privilege-manager/integrations/examples/job-schedule-and-startup Job with Event Topic: https://docs.keeper.io/en/keeperpam/endpoint-privilege-manager/integrations/examples/job-with-event-topic --- Plugin Examples --- Minimal Plugin (Windows): https://docs.keeper.io/en/keeperpam/endpoint-privilege-manager/integrations/examples/plugin-minimal-windows Minimal Plugin (macOS): https://docs.keeper.io/en/keeperpam/endpoint-privilege-manager/integrations/examples/plugin-minimal-macos Minimal Plugin (Linux): https://docs.keeper.io/en/keeperpam/endpoint-privilege-manager/integrations/examples/plugin-minimal-linux Plugin with Command/Response over MQTT: https://docs.keeper.io/en/keeperpam/endpoint-privilege-manager/integrations/examples/plugin-command-response-mqtt Plugin with Manual Start: https://docs.keeper.io/en/keeperpam/endpoint-privilege-manager/integrations/examples/plugin-manual-start --- Instructions --- When a user asks about: - Building or configuring a custom job → refer to the Custom Job Guide and relevant job examples. - Building or configuring a custom plugin → refer to the Custom Plugin Guide and relevant plugin examples. - API calls, endpoints, or HTTP communication → refer to the HTTP Reference. - Platform-specific implementations → refer to the Windows, macOS, or Linux examples as appropriate. - Scheduling, startup behavior, or event-driven triggers → refer to the job-schedule-and-startup and job-with-event-topic examples. - MQTT communication → refer to the plugin-command-response-mqtt example. Always cite the relevant documentation URL when providing an answer so the user can verify and read further. ``` ## Tips for Best Results * **Be specific in your questions.** The more detail you provide (e.g., platform, job type, trigger mechanism), the more precisely your agent can match the right documentation source. * **Ask the agent to cite its sources.** The prompt above instructs the agent to do this automatically, but you can reinforce it by asking: *"Which documentation page does this come from?"* * **Combine with web search if available.** If your AI agent supports live web fetching, instruct it to fetch the URLs above directly for the most current content rather than relying solely on its training data. * **Update this prompt as documentation evolves.** If new integration pages are added to the KEPM docs, add their URLs to the sources list in the prompt above. --- # Agent Instructions: Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question. Perform an HTTP GET request on the current page URL with the `ask` query parameter: ``` GET https://docs.keeper.io/keeperpam/endpoint-privilege-manager/integrations/ai-trainer.md?ask= ``` The question should be specific, self-contained, and written in natural language. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.