gitlab-duo-mcp-bridge
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| DUO_MOCK | No | 1/true → return a canned review (no Duo). | |
| DUO_CLI_CWD | No | Working directory Duo runs in. | |
| GITLAB_TOKEN | No | GitLab token passed through to Duo subprocess. | |
| DUO_TOOL_NAME | No | Tool name registered with MCP. | duo_review |
| DUO_TIMEOUT_MS | No | Per-run timeout (ms). | 120000 |
| DUO_CLI_COMMAND | No | Executable to launch. | glab |
| GITLAB_BASE_URL | No | GitLab base URL passed through to Duo subprocess. | |
| GITLAB_DUO_MODEL | No | Default model (e.g. gpt_5_codex). | |
| DUO_CLI_BASE_ARGS | No | Sub-command args (space separated). | duo cli run |
| DUO_CLI_GOAL_FLAG | No | Flag that carries the prompt. | --goal |
| DUO_CLI_EXTRA_ARGS | No | Extra args appended to every call. | |
| DUO_CLI_MODEL_FLAG | No | Flag that carries the model. | --model |
| DUO_MAX_OUTPUT_CHARS | No | Max raw output kept in raw. | 100000 |
| DUO_MAX_INLINE_GOAL_CHARS | No | Above this prompt size, the goal (big diffs) is sent via a temp file instead of inline. | 7000 |
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": true
} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| duo_reviewA | Run a code review with the GitLab Duo CLI and return a normalized, fault-tolerant result. Use this to get a 'second opinion' review of a diff or set of files: pass a unified |
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 1 tool
With only one tool, there is no ambiguity at all. An agent can clearly identify the single purpose.
With a single tool, there are no inconsistencies or mixing of conventions. The name 'duo_review' is clear and follows a verb_noun pattern.
The server claims to be a bridge for GitLab Duo, which likely offers multiple features, yet only one tool is provided. This under-scoping makes the count too low for the apparent domain.
The single tool covers only code review, missing other expected Duo functionalities such as code suggestions, summarization, or issue creation. This leaves significant gaps for agent workflows.