Morpheus Inference
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| MORPHEUS_AGENT | No | Alternative slot identifier (alias for MORPHEUS_AGENT_SLOT). | |
| MORPHEUS_API_KEY | No | Shared Bearer API key when MORPHEUS_KEY_MODE is 'shared'. | |
| MORPHEUS_BASE_URL | No | Custom Uplink base URL when MORPHEUS_ENDPOINT_MODE is 'custom'. Will be normalized to .../v1. | |
| MORPHEUS_KEY_MODE | No | Key mode: 'shared' for a single API key, 'per_bot' for per-bot secrets. | shared |
| MORPHEUS_AGENT_SLOT | No | Slot identifier when MORPHEUS_KEY_MODE is 'per_bot'. | |
| MORPHEUS_API_KEY_cfo | No | Per-bot API key for the 'cfo' slot. | |
| MORPHEUS_API_KEY_cio | No | Per-bot API key for the 'cio' slot. | |
| MORPHEUS_API_KEY_cmo | No | Per-bot API key for the 'cmo' slot. | |
| MORPHEUS_API_KEY_coo | No | Per-bot API key for the 'coo' slot. | |
| MORPHEUS_API_KEY_cos | No | Per-bot API key for the 'cos' slot. | |
| MORPHEUS_API_KEY_cto | No | Per-bot API key for the 'cto' slot. | |
| MORPHEUS_API_KEY_ciso | No | Per-bot API key for the 'ciso' slot. | |
| MORPHEUS_API_KEYS_JSON | No | JSON map of slot -> key for per-bot mode (escape hatch). | |
| MORPHEUS_ENDPOINT_MODE | No | Endpoint mode: 'public' for default API, 'custom' for a custom Uplink URL. | public |
| MORPHEUS_MODEL_DEFAULT | No | Default model ID used when the request omits 'model'. | |
| MORPHEUS_POLICY_MOE_DEFAULT | No | Prefer MoE via morpheus_chat (e.g., 'true' or 'false'). | |
| MORPHEUS_POLICY_CURSOR_EXCEPTION | No | Native model used only if operator names an exception. | |
| MORPHEUS_POLICY_FALLBACK_ON_DOWN | No | Fallback to native model with a reason message if Morpheus is down (e.g., 'true'). |
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 | {
"listChanged": false
} |
| prompts | {
"listChanged": false
} |
| resources | {
"subscribe": false,
"listChanged": false
} |
| experimental | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| morpheus_chatA | Call Morpheus OpenAI-compatible chat/completions. Pass messages (OpenAI shape). Model defaults to MORPHEUS_MODEL_DEFAULT when omitted. Optional tools / tool_choice for models that support tool_calls. Returns JSON with assistant content and tool_calls if any. |
| morpheus_list_modelsA | List models from GET {base}/models (install smoke + model picker). |
| morpheus_embedC | Call Morpheus OpenAI-compatible embeddings (POST {base}/embeddings). |
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 3 tools
Each tool targets a distinct operation: chat completions, embeddings, and model listing. There is no overlap or ambiguity between them.
All tools share the 'morpheus_' prefix and use a consistent verb-based naming pattern (chat, list_models, embed). The convention is uniform and predictable.
Three tools are well-scoped for an inference server providing chat and embedding endpoints plus model discovery. Each tool earns its place without bloat.
The set covers the core OpenAI-compatible inference surface: chat completions, embeddings, and model listing. No obvious dead ends or missing essential operations for this domain.