police-legal-agent
by hxyz486
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| PORT | No | HTTP service port. | 8888 |
| LAWS_DIR | No | Directory containing the law TXT files (knowledge base source). | <repository>/laws |
| CONSENSUS | No | Batch risk assessment reliability switch: dual-vote consensus. | 1 |
| LLM_MODEL | No | Text model name (can be auto-discovered). | |
| RULE_MODE | No | Rule-based fallback layer when no model is available: auto/on/off. | auto |
| FLUSH_EVERY | No | Batch risk assessment reliability switch: atomic flush interval (items). | 20 |
| LLM_API_KEY | No | API key for the text model. If not set, the service still starts: QA falls back to the deterministic rule-based synthesis layer and risk assessment returns a degraded structure. | |
| LLM_API_URL | No | Text model API URL (OpenAI-compatible gateway). | |
| LLM_RETRIES | No | Number of LLM call retries. | 3 |
| LLM_TIMEOUT | No | LLM call timeout. | 240 |
| MAX_WORKERS | No | Number of concurrent threads for batch assessment. | 4 |
| PRESET_FILE | No | Optional preset answer library; if the file is missing, normal RAG is used. | <repository>/preset/preset_answers.json |
| RETRY_FAILED | No | Batch risk assessment reliability switch: retry failed items. | 1 |
| LLM_MAX_TOKENS | No | Maximum tokens per LLM call. | 2048 |
| RERANK_API_KEY | No | API key for the rerank model. | |
| RERANK_API_URL | No | Rerank model API URL; if unavailable, reranking degrades to coarse ranking. | |
| RISK_GROUNDING | No | Switch for law-article grounding during risk assessment (0 = restore original assessment behavior). | 1 |
| VERIFY_NO_RISK | No | Batch risk assessment reliability switch: no-risk verification. | 1 |
| EMBEDDING_API_KEY | No | API key for the embedding model. | |
| EMBEDDING_API_URL | No | Embedding model API URL; if unavailable, retrieval automatically degrades to keyword search. | |
| RISK_GROUNDING_TOP_K | No | Number of grounding law articles. | 3 |
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
ActivityMaintained
ResponsivenessNo issues