cognigy

Plugin cognigy from marketplace cognigy-plugin in Cognigy/cognigy-plugin. Commands read from the repo's manifest — 2 days ago.

Skills (17)

• a2a-setup

  Use when the user explicitly asks for A2A (Agent2Agent), wants an agent to call an external / third-party A2A agent or one in another project or tenant, or wants to expose a Cognigy agent to outside A2A callers. Covers the manage_a2a_server endpoint and the create_tool { toolType: 'a2a' } tool type.

• agent-creation

  Use when the user wants to build, create, or set up a new Cognigy AI Agent from scratch, or change an agent's settings, skills, tools, knowledge or system prompt, or make it answer faster — covers checking what the environment supports, listing projects, ensuring an LLM exists, creating the agent, t

• agent-red-team

  Use when the user wants to red-team, adversarially test, jailbreak-test, or probe the guardrails of a Cognigy AI Agent — covers scoping consent, deriving the agent's policy surface from its config, generating targeted probes, scoring verdicts, and producing a standardised findings report.

• docs-lookup

  Use when the user asks how a Cognigy.AI feature, node, endpoint, setting, or concept works, asks for Cognigy documentation, best practices, or release/version behavior, or when you are unsure of platform behavior or valid configuration values and need to consult the official docs before answering or

• flow-nodes

  Use when adding, editing or debugging custom logic inside a Cognigy tool branch with manage_flow_nodes (code, HTTP Request, if/else and other nodes), or when rendering/visualizing a flow as a diagram — supported node types, config schemas, placement rules, the tool-first workflow, and the render ope

• knowledge-setup

  Use when the user wants to add knowledge, RAG, or a knowledge store/source to a Cognigy agent — covers embedding vs Knowledge Search models, Knowledge AI settings, ingesting sources, and attaching knowledge as a tool.

• llm-providers

  Use when configuring or choosing an LLM for a Cognigy agent — valid provider names (openAI, anthropic, azureOpenAI, google, mistral, openAICompatible, awsBedrock), model strings, connection types, credential resolution, OpenAI-compatible endpoints (vLLM, Hugging Face, LiteLLM, Azure AI Foundry, self

• package-management

  Use when exporting, importing, uploading, inspecting, or downloading Cognigy package zip files — including reusing an LLM plus its connection across projects.

• s2s-demo

  Use when the user wants a speech-to-speech (S2S, also called voice-to-voice or V2V) voice agent, or a callable realtime voice demo for a customer from one prompt. Covers the GPT Live voice model, the agent's voice and voice prompt, the click-to-call page, and what to do when speech-to-speech is not

• settings

  Use when configuring project-level Cognigy settings — voice preview / speech provider configuration and Knowledge AI settings.

• snapshot-backups

  Use when backing up a Cognigy project before changing an existing AI Agent, rolling a project back to a previous state, undoing agent changes, or working with Cognigy Snapshots — create, restore, list, delete.

• tools-setup

  Use when creating, updating or configuring Cognigy agent tools (create_tool, update_tool) — choosing the tool type (tool, http, mcp, knowledge, send_email, a2a) and their configuration schemas.

• troubleshooting

  Use when a Cognigy agent returns empty responses, a tool call or create_ai_agent fails, a resource is not found, setup_llm fails, or you need to diagnose a Cognigy MCP problem.

• voice-gateway-setup

  Use when the user wants to set up a voice agent (cascaded or speech-to-speech, S2S, also called voice-to-voice or V2V), choose its voice or speech provider (TTS/STT, NiCE-provided speech), change its call behaviour (greeting, barge-in, silence), or create a Voice Gateway endpoint with WebRTC for bro

• voice-go-live-checklist

  Use before calling audit_voice_agent, and whenever the user asks if a Cognigy voice agent is ready for production calls, go-live or launch, or wants it audited, validated or made production-ready. The audit covers only the API-checkable part of the Voice Go-Live Checklist; this skill lists the manua

• webchat-setup

  Use when the user wants to deploy a Cognigy agent to a Webchat endpoint — creating or updating the endpoint, style presets, layout/behavior settings, and the embed snippet.

• xapps

  Use when the user wants to add xApps to a Cognigy flow — interactive HTML / Adaptive Card mini-apps rendered to the user, with the submitted data flowing back into the conversation. Covers the xApp flow nodes, the Init Session rule, and accessing xApp data.

Subagents (3)

• cognigy-agent-builder

  Builds a new Cognigy AI Agent end to end — discovers projects, ensures a working LLM (reusing one via packages before creating a new one), creates the agent, tests it, and refines its persona/job config. Use when the user wants to create, build, or set up a new AI agent from scratch. Runs the multi-

• cognigy-s2s-demo-builder

  Builds a speech-to-speech (S2S, also called voice-to-voice or V2V) demo AI Agent for a customer end to end. It checks that speech-to-speech is enabled, reuses or creates the gpt-live-1 voice model (GPT Live), creates and tests the agent, and publishes a branded click-to-call page. Use when the user

• cognigy-voice-go-live

  Runs the Voice Go-Live Checklist audit on a Cognigy voice AI agent in an isolated context and returns a short report — what fails, what is auto-fixable, and what stays manual. Use when the user wants a voice agent production-ready, a go-live audit, or voice settings validated or fixed. Read-only unl

MCP servers (2)

• platform

  runs: npx -y -p cognigy-engine@npm:@cognigy/[email protected] cognigy-mcp

• docs

  connects to: https://docs.cognigy.com/mcp

Similar plugins