Click on "Install Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@MCP Chat chat with Claude about @architecture.md"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
MCP Chat is a command-line interface application that enables interactive chat capabilities with AI models through the Anthropic API. The application supports document retrieval, command-based prompts, and extensible tool integrations via the MCP (Model Context Protocol) architecture.
See ARCHITECTURE.md for diagrams of how main.py, mcp_client.py, mcp_server.py, and the core/ modules fit together.
Python 3.10+
Anthropic API Key
Related MCP server: MCP Chat
Copy .env.example to .env in the project root and fill in the values:
cp .env.example .envANTHROPIC_API_KEY="" # Your Anthropic API secret key
CLAUDE_MODEL="claude-sonnet-4-5" # Model id to chat with
USE_UV=1 # 1 to launch the MCP server via uv, 0 for plain pythonANTHROPIC_API_KEY and CLAUDE_MODEL are both required — the app exits at startup if either is empty.
uv is a fast Python package installer and resolver.
Install uv, if not already installed:
pip install uvCreate and activate a virtual environment:
uv venv
source .venv/bin/activate # On Windows: .venv\Scripts\activateInstall dependencies:
uv pip install -e .Run the project
uv run main.pyCreate and activate a virtual environment:
python -m venv .venv
source .venv/bin/activate # On Windows: .venv\Scripts\activateInstall dependencies:
pip install anthropic python-dotenv prompt-toolkit "mcp[cli]==1.8.0"Run the project
python main.pySimply type your message and press Enter to chat with the model.
Use the @ symbol followed by a document ID to include document content in your query:
> Tell me about @deposition.mdUse the / prefix to execute prompts defined in the MCP server. Each command takes a document id:
> /format deposition.mdCommands will auto-complete when you press Tab. mcp_server.py currently defines one prompt, format.
Edit the mcp_server.py file to add new documents to the docs dictionary.
Pass additional server scripts as arguments; each is launched as its own MCP client and its tools are merged into the tool set offered to the model:
uv run main.py path/to/other_server.pyThere are no lint or type checks wired into CI. To lint locally:
uvx ruff check .Edit a document by replacing a string in the documents content with a new string
| Name | Required | Description | Default |
|---|---|---|---|
| doc_id | Yes | Id of the document that will be edited | |
| new_str | Yes | The new text to insert in place of the old text | |
| old_str | Yes | The text to replace. Must match exactly, including whitespace |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
No annotations are provided, and the description does not disclose behavioral details such as whether all occurrences are replaced, behavior when old_str is not found, return value, or any side effects. The schema mentions exact matching, but that is in the parameter description, not the tool description.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is a single sentence with no fluff, but it contains a minor typo ('documents') and could be slightly clearer about the target content. Overall, it is concise and front-loaded.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
For a simple mutation tool, the description covers the core operation but omits important context like return value, behavior on no match, and whether replacement is global. Given no output schema, including such details would make the description more complete.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
The input schema provides full descriptions for all three parameters (doc_id, old_str, new_str) with 100% coverage. The tool description only rephrases the replace operation without adding new meaning to the parameters, so it meets the baseline for schema-covered parameters.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description uses a specific verb 'edit' with resource 'document' and clarifies the mechanism ('replacing a string with a new string'). This clearly distinguishes it from the sibling read_doc_contents.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description implies the tool is for modifying document content by string replacement, but it does not explicitly state when to use it versus reading, nor any exclusions or alternative tools. With only a read sibling, some guidance would improve clarity.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Read the contents of a document and return it as a string.
| Name | Required | Description | Default |
|---|---|---|---|
| doc_id | Yes | Id of the document to read |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations, the description carries the full burden. It discloses a key behavioral trait (returns a string), but it does not mention error handling (e.g., if doc_id is invalid), access permissions, or potential side effects (though reading implies none). This is minimal but not misleading.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is a single, front-loaded sentence with no unnecessary words. It states the action, object, and output in a compact form.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
For a simple tool with one fully documented parameter and no output schema, the description adequately states the operation and return type. It lacks edge-case details (e.g., errors or size limits), but these are not critical for basic usage, making it nearly complete.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
The schema provides 100% coverage for 'doc_id' with a description ('Id of the document to read'). The tool description adds no additional semantic information about the parameter, so the baseline of 3 is appropriate.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description uses a specific verb ('Read') with a clear resource ('contents of a document') and specifies the return type ('string'). This distinguishes it from the sibling tool 'edit_document', which implies modification rather than reading.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The usage is implied by the read vs. edit contrast with the sibling tool, but the description does not explicitly state when to use this tool over alternatives or provide any exclusions. There is no guidance on prerequisites or conditions.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Recent tool additions, removals, and schema changes observed during successful MCP inspections. Dates show when Glama detected each change.
v0.1.0edit_documentread_doc_contentsThe two tools have completely distinct purposes: one reads document content, the other edits by replacing text. There is zero overlap or ambiguity between them.
Both tools follow a verb_noun pattern (read_doc_contents, edit_document), but the noun forms differ slightly ('doc_contents' vs 'document'). This minor inconsistency doesn't hinder readability.
With only two tools, the server feels thin for a document manipulation domain. While the narrow focus is defensible, the count is below the typical 3-15 range and borders on insufficient.
The server provides read and edit (replace) operations, but lacks create, delete, list, or other common document lifecycle operations. These gaps could be worked around but represent notable missing functionality.
Unclaimed servers have limited discoverability.
If you are the server author, to access and configure the admin panel.
Talk to your public-facing AI from any MCP client — Claude, ChatGPT, Cursor, Cline, Windsurf.
Persistent memory for AI assistants. Save once; recall from Claude, ChatGPT, or any MCP client.
Real-time chat for AI agents. Claude Code, Cursor, Cline and Codex join channels over MCP.
Private persistent memory for Claude, ChatGPT & Gemini via MCP - semantic search, zero-code setup.
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/dspenard/claude_mcp_chat_cli'
If you have feedback or need assistance with the MCP directory API, please join our Discord server
