extended-rlm
by lukiies
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| MEMORY_DIR | No | Tier-1 memory search | <ws>\.kb\memory |
| RLM_CONFIG | No | Explicit config YAML path | |
| RLM_FILE_PORT | No | Stable download-server port for this agent | |
| RLM_STATS_LOG | No | Per-agent usage JSONL; its parent dir is the shared stack-stats dir | |
| READER_API_KEY | No | Reader endpoint (any OpenAI-compatible /chat/completions) | |
| RLM_AGENT_NAME | No | Agent identity; tool-name suffix + description tag (AGENT.md Domain: overrides the env domain) | |
| READER_BASE_URL | No | Reader endpoint (any OpenAI-compatible /chat/completions) | Ollama |
| RLM_AGENT_DOMAIN | No | Agent identity; tool-name suffix + description tag (AGENT.md Domain: overrides the env domain) | |
| RLM_COMMON_RULES | No | Path to the engine-wide COMMON-RULES.md served by get_session_rules | |
| RLM_READER_MODEL | No | Reader endpoint (any OpenAI-compatible /chat/completions) | qwen3:4b |
| RLM_MEMORY_SEARCH | No | Tier-1 memory search | 1 |
| RLM_READER_ENABLED | No | 0 → raw ranked chunks (no distillation) | 1 |
| RLM_READER_NOTHINK | No | Suppress Qwen3/3.5 reader thinking via assistant prefill (fast path) | |
| RLM_READER_TIMEOUT | No | Reasoning-reader headroom / HTTP timeout (s) | 120 |
| RLM_RETRIEVAL_MODE | No | grep or hybrid (opt-in embeddings: pip install -e .[embeddings]) | grep |
| KNOWLEDGE_BASE_PATH | No | Workspace root (equivalent to --path) | cwd |
| RLM_ENFORCE_SEMANTIC | No | Action-time gates | 1 |
| RLM_GROUNDING_ENABLED | No | Verdict line behaviour | 1 |
| RLM_ENFORCEMENT_ENABLED | No | Action-time gates | 1 |
| RLM_GROUNDING_STRICTNESS | No | Verdict line behaviour | strict |
| RLM_READER_THINK_ALLOWANCE | No | Reasoning-reader headroom / HTTP timeout (s) | 0 |
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
Server capabilities have not been inspected yet.
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
No tools | |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
This server cannot be deployed
Maintenance
ActivitySlowing
ResponsivenessNo issues