메뉴
Use when configuring or choosing an LLM for a Cognigy agent — valid provider names (openAI, anthropic, azureOpenAI, google, mistral), model strings, connection types, and credential resolution.
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
SOC 직업 분류 기준
Use when adding custom logic inside a Cognigy tool branch with manage_flow_nodes — supported node types, config schemas, placement rules, and the tool-first workflow.
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.
Add a new MCP tool (or extend an existing one) in the NiCE Cognigy Plugin's MCP server
Use when the user wants to build, create, or set up a new Cognigy AI Agent from scratch — covers listing projects, ensuring an LLM exists, creating the agent, testing it, and refining persona/job fields.
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.
Use when exporting, importing, uploading, inspecting, or downloading Cognigy package zip files — including reusing an LLM plus its connection across projects.
| name | llm-providers |
| description | Use when configuring or choosing an LLM for a Cognigy agent — valid provider names (openAI, anthropic, azureOpenAI, google, mistral), model strings, connection types, and credential resolution. |
| provider | modelType examples | Connection type | Notes |
|---|---|---|---|
| openAI | gpt-4o, gpt-4o-mini, gpt-4.1, gpt-4.1-mini | OpenAIProvider | Requires apiKey (sk-...) |
| anthropic | claude-sonnet-4-0, claude-opus-4-0, claude-3-opus-20240229 | AnthropicProvider | Requires apiKey |
| azureOpenAI | gpt-4o (deployment name) | AzureOpenAIProviderV2 | Requires apiKey, may need connectionId with deployment config |
| gemini-2.0-flash, gemini-1.5-pro | GoogleVertexAIProvider | Requires apiKey | |
| mistral | mistral-small-2503, mistral-medium-latest | MistralProvider | Requires apiKey |
setup_llm can create different kinds of Cognigy LLM resources. The important distinction is the modelType:
gpt-4o, gpt-4o-mini, gpt-4.1, claude-sonnet-4-0, gemini-2.0-flash, mistral-small-2503.text-embedding-3-small, text-embedding-3-large, text-embedding-ada-002, luminous-embedding-128, amazon.titan-embed-text-v2:0, Pharia-1-Embedding-4608, gemini-embedding-001, custom-embedding-model.Chat/completion models are not embedding models. gpt-4o-mini is a chat model, not a valid embedding-model choice for knowledge-store indexing.
When using list_resources { resourceType: "llm_model", projectId }, inspect the returned modelType and select the model by its exact role. Do not infer that all LLMs are interchangeable.
After creating the model, setup_llm automatically tests the connection by sending a minimal probe to the provider. This catches invalid API keys, wrong model types, and misconfigured providers before they can break downstream flows.
connectionTest.isCredentialsValid: true and the provider's confirmation message.If the connection test cannot run in your environment (e.g., air-gapped setup, unsupported custom model provider), you can pass dangerouslySkipConnectionTest: true to skip validation. This is a last resort — it may leave a non-functional model reference that silently breaks agent conversations and knowledge stores. Always prefer fixing the root cause instead. Do not use this to work around a missing, invalid, or cross-project connection.