Skip to main content
Glama
wdi-decision-intelligence

wdi-decision-intelligence

WDI — Women's Decision Intelligence MCP

An open-source Model Context Protocol server that helps Claude, Cursor or any MCP client structure high-stakes career and life decisions with numbers and evidence instead of generic reassurance: quitting a "golden handcuffs" job, taking a career break for caregiving, relocating, or weighing a family promise against what you actually control.

Zero dependencies, Python 3.10+, everything runs locally.

Tools

Tool

What it does

six_dimension_checklist

The six questions to answer for every option: cash runway, irreversible time windows, care load, safety floor (health insurance, housing), actual control rights, institutional bridges (leave, severance, part-time).

calculate_runway

Months of runway after a change, the cash needed for a safety floor (default 6 months of essentials), and how long you can experiment before touching it.

grade_evidence

Tags each claim behind the decision as KNOWN, ASSUMED, UNKNOWN or SYNTHETIC (hope or verbal promise) with a next step for each.

compare_options

Ranks 2+ options scored 1-5 on the six dimensions with optional weights. An option with a redline (score 1, e.g. losing health insurance) is ranked last regardless of its average.

Related MCP server: ideaudit-tools

Install

git clone https://github.com/alicexixi85/wdi-decision-mcp.git
{
  "mcpServers": {
    "wdi": {
      "command": "python3",
      "args": ["/absolute/path/to/wdi-decision-mcp/server.py"]
    }
  }
}

Docker: docker build -t wdi-decision-mcp . && docker run -i --rm wdi-decision-mcp

Example prompts

  • "I have $40k in savings, $3,500 of essential costs a month and would earn $1,000 freelancing. If I quit, how long is my runway with a 6-month safety floor?"

  • "Grade the evidence: my manager promised a promotion next year; the offer letter says $120k; I don't know the severance policy."

  • "Compare staying, a 4-day week, and quitting across the six dimensions."

There is also a no-install version of the method in wdi_lite_prompt.md that you can paste into any chat assistant.

Limits

This is a decision-structuring aid. It is not legal, tax, financial, medical or mental-health advice. For contracts, prenups or divorce, taxes, benefits or safety concerns, talk to a qualified professional. If you are in danger, contact local emergency services.

Full WDI toolkit

This repository is complete and free (MIT). The separate paid WDI 2.4.3 toolkit ($35, Gumroad) is a 75-file bilingual (EN/ZH) skill package for AI agents: a root SKILL file, research profiles, frameworks and playbooks, synthetic decision-card examples and jurisdiction-sensitive research cards (marked as needing verification).

License

MIT

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

    Related MCP Connectors

    Related MCP Servers

    • A
      license
      A
      quality
      B
      maintenance
      Quant tools for AI agents on Kalshi & Polymarket. 34 tools, 27 free with no key: EV, Kelly sizing, Bayes updates, odds conversion, base-rate gaps, combo grading, Fed odds, NFL and NHL models, 2026 Senate race odds, Kalshi weather board and Kalshi 15-minute markets + perps (live board + liquidation estimates). Pro adds commodity + NFL edge models, mispricing scanner and alerts.
      44
      49 npm
      MIT
    • A
      license
      B
      quality
      C
      maintenance
      Enables local deterministic scoring for startup audits with twenty pure-function tools, allowing GO/KILL verdicts, composite scores, market, money, and signal calculations without accounts, network, or model calls.
      20
      38 npm
      Apache 2.0
    • A
      license
      Not graded
      quality
      B
      maintenance
      Exposes five deterministic tools over stdio so LLM agents can evaluate all seven kill methods, plan complete well kills and CWI stripping/snubbing operations, and create or step live simulator sessions without hallucinating safety-critical pressure math. Returns kill mud weight, MAASP, stage-by-stage procedures, rig gauges, kill sheets, and graded debriefs.
      MIT
    • A
      license
      Not graded
      quality
      C
      maintenance
      Exposes four System 1 decision tools — categorical choice, rubric scoring, binary probability assertions, and a pre-execution guardrail — over stdio JSON-RPC. This enables agents to make sub-second routing, triage, and safety decisions without slow generative calls.
      MIT