Skip to main content
Tools for Democracy
ToolsThe ProblemResourcesFAQAboutRequest a Demo

Tools for Democracy

Knowledge infrastructure for governments that want to fix themselves. Built by Cory Weinstein.

Capabilities

  • Knowledge Capture
  • Confidence Monitor
  • Structured Reasoning
  • Cross-Domain Connector
  • After-Action Review
  • Policy Analyzer

Company

  • About
  • The Problem
  • All Tools
  • Resources
  • Case Studies
  • Request a Demo

Legal

  • Privacy Policy
  • Terms of Service
  • Accessibility
  • Security
© 2026 Tools for Democracy LLC. All rights reserved.

On-premise software for government agencies

I help governments keep what they know.

238,000 federal workers left in 2025. Their expertise left with them. I build software that captures institutional knowledge before it walks out the door.

Request a DemoRead: The Hidden Cost of Turnover

This is what it looks like

An analyst searches for expertise on a topic. The system returns knowledge from a colleague who retired two years ago — with confidence scores and context.

kc_search --query 'FOIA exemption 7A precedent'
Found 3 results (0.04s)

[1] Score: 0.94 | By: Maria Chen (retired 2024)
    "Exemption 7A applies when disclosure would interfere
     with enforcement proceedings. Key precedent: the 2019
     EPA case where we released partial records after the
     investigation closed. Check with OGC before applying
     to active cases — they changed the threshold in 2022."
    Tags: foia, exemption-7a, enforcement, ogc
    Confidence: 82% (last verified 2024-01-15)

[2] Score: 0.87 | By: James Wright (active)
    "7A denials require a specific, articulable harm test.
     Generic 'law enforcement purposes' language gets
     overturned on appeal. Document the actual harm."
    Confidence: 91% (verified 2025-11-02)

Maria Chen retired in 2024. Her expertise is still answering questions.

After-Action Reviews that learn

The system reads your AARs and surfaces recurring patterns your team might miss across dozens of exercises.

aar_patterns --timeframe 2024 --category 'comms-failure'
Recurring pattern detected across 4 exercises:

Pattern: "Radio interoperability failure during multi-agency response"
  Frequency: 4 of 7 exercises (57%)
  First seen: 2024-02 (Hurricane drill, Gulf Region)
  Last seen:  2024-11 (Active shooter exercise, Metro PD)

  Root causes cited:
    - Incompatible frequencies (3 exercises)
    - No pre-established common channel (2 exercises)
    - Equipment not tested before deployment (2 exercises)

  Recommended actions (from prior AARs):
    1. Establish regional common channel SOPs — cited by 2 ICs
    2. Pre-exercise radio check with all agencies
    3. Deploy mobile interop gateway (funded but unused)

  Status: UNRESOLVED — same finding in 4 consecutive exercises

Four exercises. Same failure. The software caught it. The binder on the shelf didn't.

What you get

Software, not consulting

Installable tools with APIs and search interfaces. Not a slide deck. Working software your team uses daily.

On your servers, not mine

Deploys on-premise or on your agency cloud. Air-gapped environments supported. No data leaves your network.

Starts under the micro-purchase threshold

Your team can start with a single tool on a purchase card. No RFP required. All data exports to standard JSON/CSV if you leave. No exit fees.

Six tools. License one or all.

Each runs independently on your hardware. Together, they share a knowledge layer that gets smarter over time.

Knowledge Capture

Searchable database of institutional expertise. Full-text search, contributor tracking, confidence scores.

After-Action Review

Captures lessons from operations. Matches them to future decisions automatically.

Structured Reasoning

Step-by-step decision frameworks with auditable trails. Every decision documented.

Cross-Domain Connector

Surfaces connections across siloed departments that no single team can see.

Confidence Monitor

Flags stale assessments before someone makes a decision based on outdated data.

Policy Analyzer

Scores policies on six self-correction criteria. Finds where small changes have big impact.

Each tool page includes sample output so you can see exactly what it produces. See all tools →

Who this is for

Emergency Management Directors

FEMA requires AARs after every exercise. You're still using Word templates.

Chief Knowledge Officers

One-third of your workforce is retirement-eligible. What happens when they leave?

Inspectors General

DOD has failed its financial audit 7 years running. Decision audit trails shouldn't be optional.

Legislative Research Staff

7,383 state legislators. Most have 1-3 policy staff drowning in bills.

How I'm different

Not Palantir

Palantir connects your data lakes. I preserve the expertise in your people's heads. Different problem, different tool.

Not SharePoint

SharePoint stores documents. I capture the judgment and context that makes documents useful. The “why” behind the “what.”

Not consultants

Consultants leave when the contract ends. Software stays. Your team owns the tool and the data it captures. No dependency.

The cost of doing nothing

Every month you wait, more expertise walks out the door. Here is what your agency is losing right now.

238K

federal workers left in 2025

Federal News Network

$80K

lost per analyst departure in ramp-up costs alone

42.6%

of DOE staff lost in a single year

20 of 37

GAO high-risk items trace to workforce gaps

What it costs your agency to do nothing

Conservative

$80K

1 senior analyst retires. Their 15 years of case law knowledge leaves with them. Replacement takes 18 months to reach competency.

Moderate

$400K+

5 analysts leave in a year. Six-month productivity gap per replacement. Cross-training never happened. Institutional memory has holes.

Severe

Paralysis

Department loses 40% of workforce. New hires have no one to learn from. Decisions stall. Audits fail. Mission capability degrades.

Read the full analysis →

Getting started takes 3 steps

No 18-month procurement. No committee. Start small and scale when it works.

1

Request a demo

90 minutes. Bring your real data. I show you the tools on your problem, not a canned demo.

Free. No obligation.

2

Start a $4,500 pilot

One tool. 90 days. Under the micro-purchase threshold — your team can start on a purchase card.

No RFP required.

3

Scale to your agency

Full suite under simplified acquisition. All data exports to JSON/CSV if you leave. No exit fees.

GSA Schedule eligible.

90 minutes. Your data. My tools.

Bring a real problem your team is facing. I'll analyze it live — no slides, no sales pitch — and tell you honestly whether my tools solve it. Some things they can't do.

Pilot starts at $4,500. Full tool under the micro-purchase threshold.

Request a Demo

Taking on 3 pilot agencies per quarter.