This server provides a single tool, deep_solve, that orchestrates step-by-step reasoning workflows for complex problems without calling external models. It breaks problems into ordered, dependency-based steps, runs each step in a fresh disposable workspace, and returns a compressed final answer. You can optionally enable a verification pass (verify), use compressed reasoning language (caveman mode) to keep intermendiate steps concise, and control reasoning depth with maxSteps (1–20, default 8). The server passes only compressed summaries between steps, keeping the working context small. It is stateless, requires no API keys or environment variables, and works with any MCP‑compatible client (Claude Desktop, Claude Code, Cursor, Windsurf, etc.). Note: only this reasoning tool is included—not the broader set of tools described in the README.
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., "@ephemeral-reasoning-mcp Create a step-by-step plan to learn Spanish in 6 months"
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 server that gives an LLM a disposable, step-by-step reasoning workflow, without calling any model API itself. No API key required. The host model (whatever's driving the chat - Claude, GPT, or anything else connected via MCP) does all the actual reasoning. This server just tracks the plan and hands back only compressed summaries between steps, so the working context stays small as the problem grows.
Some reasoning-pipeline MCP servers spin up their own internal LLM client (needing its own API key) to do planning/reasoning/verification behind the scenes. This one doesn't call any model at all - it's a pure state machine. The host model:
Breaks the problem into steps itself and calls containerized_reasoning_mcp_start.
Reasons through the returned step(s), using only the compressed prior summaries it's given (not full raw reasoning from earlier steps).
Calls containerized_reasoning_mcp_submit_step with a compressed summary, gets the next step(s) back.
Repeats until the plan is complete, then calls containerized_reasoning_mcp_finalize.
Because no model call happens inside the server, this works with any MCP-compatible client,
regardless of which model or provider is behind it - no ANTHROPIC_API_KEY, no provider lock-in.
The upside is zero API key / zero extra cost / works everywhere. The trade-off: since the host model does the reasoning in its own turns rather than in a truly separate hidden context, its visible output for each step still appears in the conversation transcript (there's no hiding raw reasoning traces the way a disposable sub-agent call could). What you still get is the discipline of the pipeline (ordered/parallel steps, compressed handoff between them) and no dependency on a second model or key.
Related MCP server: UltraThink
Problem -> (host plans steps) -> containerized_reasoning_mcp_start
-> Step 1 (host reasons) -> containerized_reasoning_mcp_submit_step -> Step 2 ...
-> ... -> plan complete -> containerized_reasoning_mcp_finalize -> Final Answer (+ optional verification)Steps run in the order given. Steps sharing the same parallelGroup are handed back together as
independent work the host can do in either order.
git clone https://github.com/devLlama/containerized-reasoning-mcp.git
cd containerized-reasoning-mcp
npm installThat's it - no environment variables, no API key.
npm startRuns as an MCP server over stdio.
Standard stdio MCP server - works with any client that supports MCP (Claude Desktop, Claude
Code, Cursor, Windsurf, claude.ai connectors, etc). Point the client at node plus the absolute
path to src/index.js. No env block needed.
Edit your claude_desktop_config.json (Settings -> Developer -> Edit Config):
{
"mcpServers": {
"containerized-reasoning": {
"command": "node",
"args": ["/absolute/path/to/containerized-reasoning-mcp/src/index.js"]
}
}
}Restart Claude Desktop after saving.
claude mcp add containerized-reasoning -- node /absolute/path/to/containerized-reasoning-mcp/src/index.jsOr add the same block as above to your project's .mcp.json.
Same command/args shape as above (Cursor's mcp.json uses the same schema). Consult that
app's docs for where its MCP config file lives.
claude.ai's web chat connects to remote (HTTP/SSE) MCP servers via Settings -> Connectors,
not local stdio processes launched from a browser tab. To use this server from claude.ai's web
chat specifically, you'd need to deploy it behind an HTTP/SSE MCP transport and register it as a
remote connector - running it locally via node src/index.js only works with clients that can
launch local stdio processes (Claude Desktop, Claude Code, Cursor, etc).
tool | purpose |
| Submit the problem + your own step plan; get the first step(s) back |
| Submit a step's compressed summary; get the next step(s) or a "plan complete" signal |
| Submit the final answer (+ optional self-verification); get the compiled result, closes the session |
| Inspect an in-progress session without submitting anything |
| Abandon a session |
Sessions are held in memory for the life of the server process (keyed by sessionId), so they
don't survive a server restart.
Break a problem into ordered dependency-based steps, reason through each step in a fresh disposable workspace, and return a compressed final answer.
| Name | Required | Description | Default |
|---|---|---|---|
| verify | No | Run a verification pass on the final answer (default false) | |
| caveman | No | Use compressed reasoning language inside workspaces (default false) | |
| problem | Yes | The problem or question to solve | |
| maxSteps | No | Max number of reasoning steps (default 8) |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations, the description carries the full burden of behavioral disclosure. It mentions 'fresh disposable workspace' (implying isolated, side-effect-free reasoning) and 'compressed final answer' (indicating output style), but it does not disclose potential risks like long runtime, error behavior, or whether the workspace is truly ephemeral beyond the label. This adds some context but lacks depth.
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, well-structured sentence that front-loads the core purpose and avoids redundancy. Every clause contributes meaning: 'ordered dependency-based steps', 'fresh disposable workspace', and 'compressed final answer' each add value without unnecessary filler.
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?
The tool is moderately complex (multi-step reasoning, disposable workspaces, 4 parameters) but has no output schema. The description gives a high-level process overview and the schema documents parameters, yet it does not explain what the 'compressed final answer' looks like, how verification works, or any operational limits (e.g., timeout). This leaves gaps for a tool of this complexity.
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?
Schema description coverage is 100%, with all four parameters (problem, verify, caveman, maxSteps) fully described. The tool description adds no parameter-specific information beyond the schema, so a baseline score 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 clearly states the tool's function: 'Break a problem into ordered dependency-based steps, reason through each step in a fresh disposable workspace, and return a compressed final answer.' This uses a specific verb ('break', 'reason', 'return') and resource ('problem'), and differentiates it from a generic Q&A tool by emphasizing the structured, stepwise approach.
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 provides clear context for when to use the tool: for problems that can be decomposed into ordered, dependency-based steps. It implies complex, multi-step reasoning is the intended use case. However, it does not explicitly state when not to use it or name alternatives, which is acceptable given no siblings are provided.
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.
v0.1.0deep_solveWith only one tool, there is no possibility of confusion with other tools. The tool's purpose is clearly described, making its selection unambiguous.
The single tool name 'deep_solve' is descriptive and consistent in style, even though there is no pattern to compare against. Naming is not chaotic or mixed.
A single tool is borderline thin for a reasoning server, but the tool is comprehensive and handles the full workflow. It fits the '1-2 tools feels thin' borderline category.
The tool covers the entire reasoning process from problem breakdown to step-by-step solving and final answer. No obvious gaps are apparent for the stated purpose.
Unclaimed servers have limited discoverability.
If you are the server author, to access and configure the admin panel.
AI Reasoning Cache & Consensus Layer with 11 MCP tools via Streamable HTTP.
MCP server for progressive tool usage at any scale (see https://klavis.ai)
MCP server for generating rough-draft project plans from natural-language prompts.
MCP server for building and testing AI agents with multi-model experimentation and insights.
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/devLlama/containerized-reasoning-mcp'
If you have feedback or need assistance with the MCP directory API, please join our Discord server
