{"slug": "your-agent-has-tools-does-it-have-the-right-evidence-connecting-jylus-through", "title": "Your agent has tools. Does it have the right evidence? Connecting Jylus through MCP", "summary": "Jylus, founded by an unnamed developer, has launched an evidence and state layer that connects to compatible AI agents through the Model Context Protocol (MCP), preparing source-backed context before a model reasons. The read-only MCP endpoint exposes three tools, including get_context_pack, which returns supporting source records so developers can check whether evidence establishes a claim, contains unresolved disagreements, or is missing required records. The founder recommends testing by keeping a question identical while changing one underlying record to verify the returned evidence reflects the change.", "body_md": "Your agent successfully calls three tools.\n\nThe CRM says Enterprise.\n\nBilling says the customer downgraded yesterday.\n\nThe support document describes Enterprise benefits.\n\nEvery call succeeded. The agent can still give the wrong answer.\n\nConnecting tools gives an agent access to information. Working out which facts apply together is another part of the system.\n\nI’m the founder of Jylus. We built an evidence and state layer that prepares source-backed context before a model reasons. It now connects to compatible agents through MCP.\n\nHere’s what that means for developers.\n\nWhat MCP changes\n\nModel Context Protocol gives AI applications a standard way to discover and call tools exposed by a server.\n\nFor Jylus, that means a compatible client can request evidence through the MCP connection without you writing a separate tool wrapper around each Jylus API operation.\n\nYour model handles reasoning. Jylus prepares the evidence it receives.\n\nThree tools, with distinct jobs\n\nThe connection is read-only. Your application or ingest pipeline supplies the underlying data separately.\n\nA practical first request\n\nAfter ingesting your records, ask your agent to call \"get_context_pack\" with arguments like these. Replace the namespace and dates with your own:\n\n{\n\n  \"question\": \"What is customer C-104's current plan, and which records support it?\",\n\n  \"namespace\": \"customer-events\",\n\n  \"from\": \"2026-10-01T00:00:00Z\",\n\n  \"to\": \"2026-10-08T23:59:59Z\",\n\n  \"token_budget\": 2000\n\n}\n\nThen ask it to inspect the supporting source records.\n\nDoes the evidence establish the plan? Is there an unresolved disagreement? Is a required record missing?\n\nThose are useful things to inspect before judging the final prose.\n\nConnecting it\n\nThe remote endpoint is:\n\n\"[https://api.jylus.ai/api/mcp](https://api.jylus.ai/api/mcp)\"\n\nIt uses Streamable HTTP and a workspace API key with \"events:read\". Keep the key in private client settings or a secret store.\n\nClient-specific setup:\n\n[https://jylus.ai/mcp](https://jylus.ai/mcp)\n\nConnecting the server does not automatically capture your chat history or upload your project. It gives the client tools to query data already retained in your Jylus workspace.\n\nThe test I’d start with\n\nKeep the question identical. Change one underlying record.\n\nAsk again.\n\nCheck whether the returned evidence reflects the change, whether the supporting source IDs make sense, and whether incomplete evidence remains visible to the model.\n\nYou can explore the Context Packs first in the browser playground:\n\n[https://jylus.ai/try](https://jylus.ai/try)\n\nFor developers using MCP: what takes more work today—connecting another tool, or making the information from your existing tools agree?", "url": "https://wpnews.pro/news/your-agent-has-tools-does-it-have-the-right-evidence-connecting-jylus-through", "canonical_source": "https://dev.to/josh_hodgetts_f9f91ff3a23/your-agent-has-tools-does-it-have-the-right-evidence-connecting-jylus-through-mcp-4c3g", "published_at": "2026-10-08 11:45:12+00:00", "updated_at": "2026-10-08 11:48:50.502051+00:00", "lang": "en", "topics": ["ai-agents", "agent-protocols", "ai-tools", "developer-tools", "artificial-intelligence"], "entities": ["Jylus", "Model Context Protocol", "MCP"], "also_reported_by": [], "alternates": {"html": "https://wpnews.pro/news/your-agent-has-tools-does-it-have-the-right-evidence-connecting-jylus-through", "markdown": "https://wpnews.pro/news/your-agent-has-tools-does-it-have-the-right-evidence-connecting-jylus-through.md", "text": "https://wpnews.pro/news/your-agent-has-tools-does-it-have-the-right-evidence-connecting-jylus-through.txt", "jsonld": "https://wpnews.pro/news/your-agent-has-tools-does-it-have-the-right-evidence-connecting-jylus-through.jsonld"}}