Converse MCP Server
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| PORT | No | Server port | 3157 |
| LOG_LEVEL | No | Logging level | info |
| XAI_API_KEY | No | X.AI API key | |
| CODEX_API_KEY | No | Codex API key (optional if ChatGPT login available) | |
| GEMINI_API_KEY | No | Google Gemini API key (takes priority over GOOGLE_API_KEY) | |
| GOOGLE_API_KEY | No | Alternative Google API key | |
| OPENAI_API_KEY | No | OpenAI API key | |
| MISTRAL_API_KEY | No | Mistral API key | |
| DEEPSEEK_API_KEY | No | DeepSeek API key | |
| MCP_TOOL_TIMEOUT | No | Tool timeout in ms (system-wide) | |
| OPENROUTER_TITLE | No | OpenRouter title | |
| ANTHROPIC_API_KEY | No | Anthropic API key | |
| CODEX_SANDBOX_MODE | No | Codex sandbox mode | read-only |
| OPENROUTER_API_KEY | No | OpenRouter API key | |
| OPENROUTER_REFERER | No | OpenRouter referer | |
| SUMMARIZATION_MODEL | No | Model for summarization | gpt-5-nano |
| CODEX_SKIP_GIT_CHECK | No | Skip git check | true |
| CODEX_APPROVAL_POLICY | No | Codex approval policy | never |
| MAX_MCP_OUTPUT_TOKENS | No | Max output tokens (system-wide) | |
| OPENROUTER_DYNAMIC_MODELS | No | Enable dynamic models | true |
| ENABLE_RESPONSE_SUMMARIZATION | No | Enable AI-generated summaries | false |
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 | {} |
| prompts | {} |
| resources | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| chatA | UNIFIED CHAT — talk to one or more AI models. mode "chat" (default): 1..N models answer independently in parallel. mode "consensus": ≥2 models answer, then refine after seeing each other. mode "roundtable": models answer sequentially, each building on the running transcript. Supports files, images, and continuation_id for multi-turn threads (you may switch modes on resume). Use model "auto" for automatic selection. IMPORTANT: use the "files" parameter to share code/file content instead of pasting into the prompt. |
| check_statusA | Check the status and progress of async jobs. Query specific jobs by continuation_id or list the 10 most recent jobs. Returns job status with start time and progress information. |
| cancel_jobA | Cancel a running async job by its continuation_id. Terminates queued or running jobs with graceful cleanup. Preserves partial results when available. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
| help | Comprehensive guide for Converse MCP Server - tools, models, parameters, and configuration |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
| Help Documentation | Comprehensive guide for the Converse MCP Server including all tools, parameters, providers, and models |
TDQS
Scored across 3 tools
chat, check_status, and cancel_job have clearly distinct roles: starting/resuming conversations, inspecting job progress, and canceling jobs. There is no meaningful overlap or ambiguity between any pair of tools.
check_status and cancel_job follow a clean verb_noun pattern, and chat is a simple imperative verb that fits the conversational focus. The slight deviation from the verb_noun style is minor and does not hurt usability.
Three tools is a minimal but well-scoped set for a chat-focused server. Each tool earns its place: the main chat capability plus the two async-job management tools needed to support long-running conversations.
The core workflow of starting a chat, checking async job status, and canceling jobs is covered. The main gap is the lack of a dedicated result-fetching tool if completed outputs are not returned directly by chat or check_status.