Analysis

Provides a FastMCP server for critical thinking and multi-perspective analysis of current affairs.

python

0

GitHub Stars

python

Language

7 months ago

First Indexed

3 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
  "mcpServers": {
    "rcsnyder-analysis_mcp": {
      "command": "uvx",
      "args": [
        "git+https://github.com/YOUR_USERNAME/analysis_mcp",
        "analysis-mcp"
      ],
      "env": {
        "PYTHONPATH": "/absolute/path/to/analysis_mcp/src"
      }
    }
  }
}

You run a FastMCP server dedicated to analyzing current affairs through multi-perspective, chained thinking. It lets you deconstruct claims, explore assumptions, and iteratively build deeper insights with prompt chaining, all while remaining self-contained.

How to use

You integrate the server with your MCP client to perform iterative analyses. Start with a claim, then progressively apply analytical operations to reveal structure, assumptions, tensions, and the strongest form of the claim. Use the suggested next-step prompts to guide subsequent analysis and maintain a clear trace of every step.

How to install

Prerequisites you need before starting:

Python (for local development) and a virtual environment tool (venv)
uvx (for Quick Start with Claude Desktop)

Step-by-step setup options you can follow, depending on how you prefer to run the server.

Option A: Quick Start with Claude Desktop (uvx) using the provided MCP entry

Additional setup steps for local development

Clone the project, set up a virtual environment, install dependencies in editable mode, run tests, and start the server for testing.

Configuration and running notes

Two explicit MCP runtime configurations are shown for local usage. The first uses uvx to launch the server via a Git URL and the analysis-mcp identifier. The second runs the Python module directly, with a working directory and Python path prepared for development.

Troubleshooting

If the server isn’t connecting, ensure your runtime tool (uvx) is installed and that your config entry is valid. If tools do not appear in your client, restart the client after updating the MCP configuration and confirm the connection indicator shows the server as connected.

Notes

The server supports chain-based analysis with a suite of tools for deconstruction, synthesis, and meta-analysis. It is designed to be provider-agnostic and self-guided, with full traceability of each step.

Available tools

deconstruct_claim

Break down claims into components such as claim, evidence, implications, and assumptions.

chain_analysis

Apply analytical operations in sequence to build layered insights from previous outputs.

apply_lens

Analyze content through predefined lenses (historical, economic, geopolitical, etc.).

get_trace

Retrieve stored analysis plans and trace data by trace_id.

apply_operation

Apply 15+ analytical operations to content to derive structured insights.

list_available_operations

Show all analytical operations available for use.

extract_assumptions

Identify implicit and explicit assumptions within a claim or analysis.

identify_contradictions

Find logical tensions or conflicting positions in the analysis.

find_fallacies

Detect rhetorical or logical fallacies in arguments.

steelman_argument

Construct the strongest defensible version of a claim.

find_analogies

Identify relevant precedents or analogies to illuminate the issue.

extract_principles

Derive universal patterns or guiding principles from the analysis.

synthesize_perspectives

Merge multiple viewpoints into a coherent, integrated view.

elevate_abstraction

Raise the discussion to higher-level, systemic concepts.

ground_in_specifics

Add concrete examples to anchor abstract analysis.

identify_gaps

Identify missing elements or areas needing further exploration.

check_coherence

Verify logical consistency across the analysis.

suggest_next_step

Recommend the next operation to continue the analysis.

convert_to_dialogue

Reframe the content as a Socratic dialogue.

extract_counterfactuals

Generate alternate what-if scenarios to test robustness.

map_dependencies

Chart logical dependencies between claims, evidence, and conclusions.