Skill Claude CodeCodex
Use whenever adding a new Lyzr API endpoint (or small batch) as an MCP tool in this server, or wiring a new Lyzr service client into it.
Skill Claude CodeCodex
Use whenever adding a new Lyzr API endpoint (or small batch) as an MCP tool in this server, or wiring a new Lyzr service client into it.
Skill Claude CodeCodex
Load this skill before creating, configuring, updating, deleting, or chatting with a Lyzr agent via the lyzr MCP server, or before starting/polling a long-running agent task. Covers: choosing a provider/model, generating and reusing sessionid across a conversation, when to use lyzrchat vs lyzrstreamchat vs…
Skill Claude CodeCodex
Load this skill before creating, listing, updating, or deleting a stored API credential, or before adding/syncing/pausing/resuming a live source that keeps a knowledge base continuously synced to an external system, via the lyzr MCP server. Covers not echoing raw secrets back into chat, and the three-layer live-source…
Skill Claude CodeCodex
Load this skill before parsing or training a knowledge base from an actual binary file (PDF, DOCX, TXT, CSV, XLSX, PPTX, image) via the lyzr MCP server, or before extracting structured fields from a document. Covers the base64-encode-the-file-yourself pattern every one of these tools requires, parse-vs-train tool…
Skill Claude CodeCodex
Load this skill before creating, listing, fetching, or deleting a Responsible AI (RAI) guardrail policy via the lyzr MCP server. Covers passing flat inputs (toxicitythreshold, bannedtopics, etc.) rather than constructing the nested API shape by hand, checking existing policies before creating a duplicate, and…
Skill Claude CodeCodex
Load this skill before connecting a knowledge base to an external system (SharePoint, a shared drive) for continuous syncing via the lyzr MCP server's KB Sync connectors, or before checking/pausing/resuming an existing sync's status and attempt history. Covers the three-layer connector/credential/cc-pair model, why…
Skill Claude CodeCodex
Load this skill before creating a RAG knowledge base, training it on text/website/document content, querying it, or managing (listing, filtering, bulk-deleting) its indexed documents via the lyzr MCP server. Also load it when the user wants standalone text/website parsing or classification without a knowledge base.…
Skill Claude CodeCodex
Load this skill before building or querying a knowledge graph (entities + relationships, not chunked retrieval text) from text, a website, or a file via the lyzr MCP server — including the Neo4j-backed variants. Covers the default-store vs Neo4j-store distinction, sync vs async/task training, deduplicating entities…
Skill Claude CodeCodex
Load this skill before storing, searching, listing, updating, or deleting long-term Cognis memories via the lyzr MCP server. Covers scoping memories correctly by ownerid/userid/agentid/sessionid to avoid leaking one user's facts into another's results, and distinguishing Cognis memory from the separate Knowledge Graph…
Skill Claude CodeCodex
Load this skill before creating, listing, pausing, resuming, triggering, or deleting a cron schedule that runs a Lyzr agent automatically via the lyzr MCP server. Covers cronexpression/timezone/retry defaults, using an immediate trigger to test before leaving a schedule unattended, and preferring pause/resume over…
MCP server Claude CodeCodexCursor +2
MCP server wrapping the Lyzr Enterprise Agent API (bring-your-own-key). Runs locally from the lyzr-mcp npm package. Needs 1 environment variable to run.