generated: '2026-09-05'
method: generated
source: >-
  openapi/cognizant-technology-neuro-san-agent-service.json + grpc/*.proto +
  conventions/cognizant-technology-conventions.yml + the neuro-san documentation
note: >-
  Packaged Agent Skills for the Cognizant Neuro SAN agent service. Every operationId
  referenced was grepped out of the published OpenAPI before being written down; none is
  invented. The provider publishes no skills/ or AGENTS.md of its own — searched the neuro-san
  repository and found none — so these are generated from the contract and the provider's
  documentation.
  Two skills cover the whole surface: the four operations split cleanly into discovery (three)
  and conversation (one).
skills:
- name: cognizant-technology-discover-agent-networks
  file: cognizant-technology-discover-agent-networks.md
  description: >-
    Discover which agent networks a deployment exposes, what each does, and what it needs —
    before starting a conversation.
  operations:
  - ConciergeService_List
  - AgentService_Function
  - AgentService_Connectivity
- name: cognizant-technology-converse-with-an-agent-network
  file: cognizant-technology-converse-with-an-agent-network.md
  description: >-
    Hold a multi-turn conversation over the streaming HTTP API — reading the stream, carrying
    chat_context forward, and keeping private data on the sly_data channel.
  operations:
  - AgentService_StreamingChat
  - AgentService_Function
coverage:
  operations_in_spec: 4
  operations_covered: 4