System-1 Gateway
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| SYSTEM1_CONFIG | No | Path to the gateway config file. | |
| SYSTEM1_FALLBACK | No | Comma-separated list of fallback providers tried on failure. | |
| SYSTEM1_LOG_FILE | No | Optional file path to also write logs to. Logs always go to stderr. | |
| SYSTEM1_PROVIDER | No | Primary decision provider. Defaults to "laya". | |
| SYSTEM1_LOG_LEVEL | No | Log level, e.g. INFO or DEBUG. Defaults to INFO. | |
| SYSTEM1_TIMEOUT_S | No | Timeout in seconds per provider attempt. Defaults to 5.0. | |
| SYSTEM1_VON_MODEL | No | Von model name. Defaults to von-1.3. | |
| SYSTEM1_LAYA_MODEL | No | Pinned Laya checkpoint/model name, e.g. typed-decisions. | |
| SYSTEM1_SKILLS_HINT | No | Enables or disables skill-picker hints. Defaults to false. | |
| SYSTEM1_LAYA_API_KEY | No | API key for laya-serve if it is running with LAYA_API_KEY. | |
| SYSTEM1_OLLAMA_MODEL | No | Ollama model name, e.g. gemma4:e4b. | |
| SYSTEM1_VON_ENDPOINT | No | Von provider endpoint. Defaults to http://127.0.0.1:8777. | |
| SYSTEM1_LAYA_ENDPOINT | No | Laya provider endpoint. Defaults to http://127.0.0.1:8765. | |
| SYSTEM1_MIN_CONFIDENCE | No | Minimum confidence before abstaining. Defaults to 0.6. | |
| SYSTEM1_SKILLS_ENABLED | No | Enables or disables the skill picker. Defaults to false. | |
| SYSTEM1_OLLAMA_ENDPOINT | No | Ollama provider endpoint. Defaults to http://127.0.0.1:11434. | |
| SYSTEM1_LAYA_ALLOW_REMOTE | No | Set to true to allow non-loopback Laya hosts. Defaults to false. | |
| SYSTEM1_SKILLS_MIN_CONFIDENCE | No | Minimum confidence for skill-picker hints. Defaults to 0.8. |
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
} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| system1_decideA | Ask the configured System-1 provider for a fast decision. Returns status decided | abstained | error, with decision, confidence (0-1), rationale and request_id. |
| system1_healthA | Check gateway and provider health (ok | degraded | down). |
| system1_capabilitiesA | Describe the gateway: version, schema versions, provider chain, thresholds and JSON schemas. |
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 clearly distinct concern: making a decision, checking health, and introspecting capabilities. There is no overlap in purpose or resource, so an agent can select without confusion.
All three tools share the system1_ prefix and snake_case verb/noun form (decide, health, capabilities). The pattern is predictable and uniform throughout.
Three tools is on the lean side, but for a thin inference gateway (decide + health + capabilities) each earns its place. Slightly under-scoped if richer workflows are expected.
The core lifecycle of a decision gateway is covered: decision, health monitoring, and capability discovery. Minor gaps exist, such as batch decisions or recording decision outcomes, but agents can work around them.