NyxID
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
No arguments | |||
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| nyx_proxy_requestA | Forward an HTTP request through NyxID to any downstream service the agent has been granted access to. NyxID injects the appropriate credential at proxy time so the agent never holds raw API keys. Supports cloud APIs (OpenAI, GitHub, Slack, Lark, Telegram, etc.), internal REST endpoints, and localhost services reached over a NAT-pierced credential node. Every call is rate-limited and audit-logged per agent identity. |
| nyx_list_servicesA | Enumerate the downstream services this agent can call through NyxID. Returns each service's slug, display name, base URL, auth method, configured rate limits, and (when an OpenAPI spec is available) its callable endpoints. Use this for tool discovery before issuing nyx_proxy_request — agent-key scope determines which services are visible. |
| nyx_request_approvalA | Request human approval before performing a sensitive action. NyxID delivers the request to the user via push notification (Telegram or mobile app) and blocks the agent until the user approves, denies, or the request times out. Use this for destructive operations, financial transactions, or any action where unattended automation is inappropriate. Approvals can be granted ad-hoc per call or via pre-configured grant rules. |
| nyx_exchange_identityA | Exchange the agent's NyxID identity for a delegated access token bound to a downstream OIDC service (RFC 8693 token exchange). Lets the agent call APIs on behalf of a specific user without holding that user's long-lived credentials. The returned token carries the original user's identity claims while remaining auditable as an agent action. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 4 tools
Each tool addresses a distinct function: token exchange, service discovery, proxied requests, and human approval. No two tools could be confused for the same purpose.
All tools follow the nyx_verb_noun pattern consistently (exchange_identity, list_services, proxy_request, request_approval). No mixing of styles.
Four tools is well-scoped for an identity-and-proxy server; each tool fills a necessary role without redundancy or excessive granularity.
The set covers the core lifecycle: identity exchange, service discovery, proxy calling, and human approvals. Minor gaps like token revocation or listing active approvals exist but are workable.