ultimate-debate-mcp
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| CACHE_TTL_MS | No | Cache TTL (1 hour) | 3600000 |
| OPENAI_MODEL | No | Model to use (gpt-5.2, gpt-4o, etc.) | gpt-5.2 |
| CACHE_MAX_SIZE | No | Max cached responses | 100 |
| OPENAI_API_KEY | Yes | Your OpenAI API key | |
| RATE_LIMIT_RPH | No | Max requests per hour | 1000 |
| RATE_LIMIT_RPM | No | Max requests per minute | 100 |
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 | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| debateA | Runs a full multi-round debate loop where ChatGPT critiques content and suggests revisions. Returns structured issues, patches, and test recommendations. |
| gpt_critiqueB | Single-pass adversarial critique of content using ChatGPT. Identifies logical gaps, edge cases, security issues, and incorrect assumptions. |
| gpt_verifyA | Strict verification that previously identified issues have been resolved in revised content. Also detects any new issues introduced. |
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
gpt_critique and debate both involve critique, so there is some boundary overlap, but the descriptions clearly separate single-pass critique from a full multi-round debate loop. gpt_verify is distinctly focused on checking whether prior issues were resolved, making misselection unlikely.
Two tools follow a gpt_<verb> pattern (gpt_critique, gpt_verify), while debate breaks the prefix pattern but still uses a single clear lowercase verb/noun. The naming is generally consistent and readable, with only the missing prefix creating minor inconsistency.
Three tools is a tight, well-scoped set for a debate-focused server. Each tool covers a distinct mode—single critique, verification of revisions, and full multi-round debate—so none feel redundant or unnecessary.
The surface covers the core debate workflow: critique content, run a full debate loop with revisions and patch suggestions, and verify that issues are resolved. There are no obvious dead ends for an agent trying to critique, revise, and validate content.