Grammarly MCP Server
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| LOG_LEVEL | No | Logging level: 'debug', 'info', 'warn', 'error' | info |
| CLAUDE_MODEL | No | Claude model: 'auto', 'haiku', 'sonnet', 'opus'. Auto-selects based on text length and iteration count. | auto |
| GOOGLE_MODEL | No | Google/Gemini model name | gemini-2.5-flash |
| OPENAI_MODEL | No | OpenAI model name | gpt-4o |
| CLAUDE_API_KEY | No | Claude API key. If not set, uses claude login CLI auth | |
| GEMINI_API_KEY | No | Alternative to GOOGLE_GENERATIVE_AI_API_KEY. Google API key for Gemini models. | |
| OPENAI_API_KEY | No | OpenAI API key. Required when using OpenAI provider. | |
| ANTHROPIC_MODEL | No | Anthropic model id when using direct Anthropic provider | claude-sonnet-4-20250514 |
| STAGEHAND_MODEL | No | Deprecated. Use STAGEHAND_LLM_PROVIDER + model vars instead | |
| BROWSER_PROVIDER | No | Browser automation provider: 'stagehand' or 'browser-use' | stagehand |
| ANTHROPIC_API_KEY | No | Anthropic API key. Required when using direct Anthropic provider. | |
| IGNORE_SYSTEM_ENV | No | When true, ignores shell env vars and uses only .env file. Prevents IDE-inherited env pollution. | false |
| CONNECT_TIMEOUT_MS | No | MCP connection timeout in milliseconds | 30000 |
| BROWSERBASE_API_KEY | No | API key from browserbase.com (required for Stagehand provider) | |
| BROWSER_USE_API_KEY | No | API key from cloud.browser-use.com (required for Browser Use provider) | |
| STAGEHAND_CACHE_DIR | No | Directory for action caching | |
| REWRITE_LLM_PROVIDER | No | LLM for text rewriting: 'claude-code', 'openai', 'google', 'anthropic'. Auto-detects if not set. | |
| BROWSERBASE_CONTEXT_ID | No | Persistent context for Grammarly login state | |
| BROWSERBASE_PROJECT_ID | No | Project ID from Browserbase dashboard (required for Stagehand provider) | |
| BROWSERBASE_SESSION_ID | No | Reuse existing session (advanced) | |
| BROWSER_USE_PROFILE_ID | No | Profile with synced Grammarly login (required for Browser Use provider) | |
| LLM_REQUEST_TIMEOUT_MS | No | LLM request timeout in milliseconds | 120000 |
| STAGEHAND_LLM_PROVIDER | No | LLM for browser automation: 'claude-code', 'openai', 'google', 'anthropic'. Auto-detects if not set. | |
| CLAUDE_REQUEST_TIMEOUT_MS | No | Legacy timeout setting. Use LLM_REQUEST_TIMEOUT_MS instead. | |
| GOOGLE_GENERATIVE_AI_API_KEY | No | Google API key for Gemini models. Also accepts GEMINI_API_KEY. Required when using Google/Gemini models (the default). |
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 |
|---|---|
| grammarly_optimize_textA | Get AI detection and plagiarism scores from Grammarly, optionally rewriting text with the configured LLM to reduce scores. Modes: 'score_only' (scores only), 'analyze' (scores + recommendations), 'optimize' (iterative rewriting to meet thresholds). |
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 possibility of ambiguity or overlap between tools. The tool's purpose is clearly defined and distinct by default.
A single tool inherently has perfect naming consistency, as there are no other tools to compare against. The name 'grammarly_optimize_text' follows a clear verb_noun pattern.
One tool is too few for a server named 'Grammarly MCP Server', which implies broader functionality like grammar checking, style suggestions, or tone analysis. The single tool focuses narrowly on optimization and scoring, leaving obvious gaps in the domain.
The tool set is severely incomplete for a Grammarly-like service. It lacks basic operations such as grammar checking, spell checking, readability analysis, or tone adjustments, which are core to the implied domain. The single tool offers only optimization and scoring, creating significant gaps.