# SketricGen MCP for AI Agents AI Agent Connect > SketricGen connects your AI agents to a high-level orchestration layer. It lets you run multi-agent workflows, build searchable knowledge bases, and track every step of an execution trace. Use it to manage CRM-style contact profiles and maintain conversation history across complex tasks. ## Overview - **Category:** developer-tools - **Price:** Free - **Endpoint:** https://edge.vinkius.com/vk_preview_1oQD0BGOH8qoRrYb4Ucw7jvZh5NHJimdTtL2Lk2H/ai-agent-connect - **Tags:** workflow-automation, multi-agent-systems, knowledge-base, execution-tracing, ai-templates, credit-tracking ## Description Imagine you're trying to build a system where multiple AI agents need to talk to each other to solve a complex problem. Usually, this is a mess of nested loops and manual prompts that break as soon as the input changes slightly. SketricGen changes that by giving your agent a way to trigger these workflows programmatically. You can feed it specific contact info from a CRM and have it query a knowledge base that actually stays updated with your latest data. It handles the heavy lifting of keeping track of who said what and how much credit you're burning on each run. If something goes sideways, you don't have to guess where the logic broke or which step caused the hallucination. You can look at the exact path the agent took, including every tool call and response. It's a lot easier to manage when you're using a platform like Vinkius to keep everything organized in one place. You get a clear view of your agent's actions without having to dig through raw logs every five minutes. By centralizing these actions, you can build out a production environment where your agents actually follow rules instead of just guessing the next word. This means you spend less time debugging failed loops and more time actually shipping features. ## Tools ### get_agent Get the specific details for one AI agent. Use this to check the configuration of a specific agent. ### get_contact Get the profile details for a specific contact. This lets you pull specific customer info into your agent's context. ### get_conversation Fetch a specific conversation history. This helps you see what the agent and user discussed previously. ### get_knowledge_base Retrieve details about a specific knowledge base. Use this to see what data your agent has access to for specific tasks. ### get_trace_credits Check how many credits were used during a trace. This helps you monitor your costs and stay within your budget. ### get_trace Look at the full execution details for a specific run. You can use this to see exactly where a workflow failed or succeeded. ### get_workflow Get the details for a specific workflow configuration. This lets you check the structure of your multi-agent logic. ### list_agents See all the AI agents currently in your account. This is the quickest way to see your active agent roster. ### list_conversations See a list of all previous conversations. This is useful for auditing past interactions with your agents. ### list_contacts Browse all CRM-style contact profiles. You can use this to see your customer base or lead list. ### list_knowledge_bases View all your searchable knowledge bases. This helps you manage different data sources for different agents. ### list_templates See the available workflow templates. Use this to find inspiration or pre-made logic for your next project. ### list_traces See a list of all recent execution traces. This lets you quickly find specific runs to debug or review. ### list_workflows View all the workflows you have created. This helps you keep track of all your automated logic. ### run_workflow_with_contact Trigger a workflow specifically for a saved contact. This allows your agent to perform actions tailored to a specific customer. ### run_workflow Start a multi-agent workflow execution. Use this to kick off a complex sequence of agent actions. ### check_sketricgen_status Verify that your connection to the platform is active. Use this to ensure your API key is working correctly. ### delete_conversation Remove a conversation from your history. This lets you clean up old data or sensitive interactions. ## Prompt Examples **Prompt:** ``` Run the customer support workflow for the user 'John Doe' who is asking about a refund. ``` **Response:** ``` I've initiated the refund workflow for **John Doe**. **Action Summary:** * **Contact Located:** Found profile in CRM. * **Knowledge Check:** Queried 'Refund Policy' in the Support Knowledge Base. * **Status:** Agent is currently verifying the refund eligibility based on the 30-day policy. Would you like me to notify the user once the verification is complete? ``` **Prompt:** ``` Show me the trace for the last workflow run and tell me how many credits it used. ``` **Response:** ``` Here is the summary for the last execution: | Metric | Value | | :--- | :--- | | **Duration** | 4.8 seconds | | **Credits Used** | 12 | | **Status** | Success | **Execution Path:** 1. `get_contact` -> Success 2. `query_knowledge_base` -> Success 3. `generate_response` -> Success Do you want to see the full raw trace for these steps? ``` **Prompt:** ``` List my knowledge bases and tell me which ones are connected to my support agents. ``` **Response:** ``` You have **3** active knowledge bases: * **Internal HR Docs** (Connected to: HR Assistant) * **Public FAQs** (Connected to: Support Bot, Sales Bot) * **Technical Specs** (Connected to: Engineering Agent) The 'Public FAQs' base currently contains 142 indexed documents. ``` ## Capabilities ### Run multi-agent workflows Trigger complex sequences of agent actions programmatically. ### Query searchable knowledge bases Pull relevant info from your custom data sources in real time. ### Trace execution steps See exactly how an agent reached a specific conclusion. ### Manage contact profiles Keep CRM-style data for your users and contacts. ### Track credit consumption Monitor costs and usage per workflow run. ## Use Cases ### Automated Customer Support An agent uses `run_workflow_with_contact` to handle a ticket, pulling specific user history and documentation from a knowledge base. ### Internal HR Querying A user asks about company benefits, and the agent pulls from the HR knowledge base to give a precise, grounded answer. ### Sales Lead Management An agent checks `list_contacts` to see if a lead exists before running a specific outreach workflow to avoid duplicates. ### Production Debugging An engineer uses `list_traces` to find a failed run and then `get_trace` to see the exact tool call that caused the error. ## Benefits - You can see exactly where a multi-agent workflow failed by using `get_trace` to inspect every step. - Your agents stay grounded in your specific data by querying `list_knowledge_bases` and their contents. - You keep your CRM data organized by managing profiles with `list_contacts` and `get_contact`. - You can monitor your spending and usage by checking `get_trace_credits` for every run. - You can quickly deploy new logic by using `list_templates` to find pre-made workflow structures. ## How It Works The bottom line is you get a production-ready orchestration layer for complex agentic tasks. 1. Subscribe to the SketricGen MCP on Vinkius. 2. Enter your SketricGen API Key into your client settings. 3. Start triggering workflows and managing agents from your chat. ## Frequently Asked Questions **What is SketricGen MCP for?** SketricGen MCP is used for orchestrating multi-agent workflows and managing searchable knowledge bases. It helps your AI agents perform complex, multi-step tasks with full audit trails. **Can SketricGen help with my CRM data?** Yes. It manages contact profiles directly, allowing your agents to pull customer information into their context to provide more personalized responses. **How do I see why my agent failed?** You can use the tracing feature to see exactly where a workflow failed. It shows every step the agent took, including tool calls and the logic used at each point. **Does SketricGen support knowledge bases?** Yes, it allows you to create and query vector-searchable knowledge bases. This ensures your agents stay grounded in your specific company data. **Can I see how much my workflows cost?** Yes. SketricGen tracks credit consumption for every run, so you can monitor your usage and manage your costs effectively. **How do I run a specific workflow for a customer?** You can trigger a workflow specifically for a saved contact. This allows the agent to perform actions tailored to a specific customer's profile and history. **Can I see my past conversations?** Yes, SketricGen allows you to list and retrieve previous conversation histories, which is helpful for auditing past interactions or maintaining context. **How can my AI analyze the execution steps of a specific workflow?** Simply use the `debug_workflow_trace` tool. Your agent will instantly retrieve the execution trace, detailing every tool call, data transfer between agents, and total credit consumption per run. **Is it possible to programmatically manage and query documents for my agents?** Yes. By executing the `manage_knowledge_base` action, your AI agent can upload files (PDF, DOCX, TXT, HTML) or search existing repositories to provide context-aware, highly accurate responses. **Can I automatically trigger workflows tied to specific customer records?** Absolutely. Ask the agent to use the `run_workflow_for_contact` tool. It will execute the designated workflow while injecting the specific contact's history, enabling personalized, data-driven outputs.