Tru8 Evidence Research

Official registry Search and web

Source-traced evidence research for AI agents. We organise the evidence; you decide.

Install

Add this to your client's MCP configuration (Claude Desktop: claude_desktop_config.json; Cursor: .cursor/mcp.json; VS Code and others use the same "mcpServers" shape).

Remote (streamable-http)
Claude Code
claude mcp add --transport http tru8-evidence-research 'https://api.trueight.com/mcp/' --header 'X-API-Key: <Tru8 API key (tru8_sk_...). Create at dashboard > Settings > Developer.>'
JSON configuration (Claude Desktop, Cursor, VS Code)
{
    "mcpServers": {
        "tru8-evidence-research": {
            "type": "http",
            "url": "https://api.trueight.com/mcp/",
            "headers": {
                "X-API-Key": "<Tru8 API key (tru8_sk_...). Create at dashboard > Settings > Developer.>"
            }
        }
    }
}
PYPI package tru8-mcp
Claude Code
claude mcp add tru8-evidence-research -e 'TRU8_API_KEY=<your tru8_api_key>' -e 'TRU8_API_URL=<TRU8_API_URL>' -- python -m tru8_mcp tru8-mcp==1.0.3
JSON configuration (Claude Desktop, Cursor, VS Code)
{
    "mcpServers": {
        "tru8-evidence-research": {
            "command": "python -m tru8_mcp",
            "args": [
                "tru8-mcp==1.0.3"
            ],
            "env": {
                "TRU8_API_KEY": "<your tru8_api_key>",
                "TRU8_API_URL": "<TRU8_API_URL>"
            }
        }
    }
}

Environment variables

  • TRU8_API_KEY required secret Tru8 API key (tru8_sk_...). Create at dashboard Settings > Developer. Store in env vars, never hardcode.
  • TRU8_API_URL API base URL. Default: https://api.trueight.com

Documentation

From the project's README on GitHub, refreshed weekly. Links open on GitHub.

Contents
  1. Tru8 — AI-Powered Evidence Research Platform
  2. Quick Start
  3. Project Structure
  4. Pipeline
  5. Six Evidence Views
  6. API & Agent Access
  7. Testing
  8. Infrastructure
  9. API Documentation
  10. Environment Variables
  11. Licence

Tru8 — AI-Powered Evidence Research Platform

Tru8 organises evidence so you can decide what matters. Submit a claim or URL, and Tru8 retrieves evidence from the open web and specialist APIs, classifies it by tier (primary / reporting / commentary) and type (data / official / news / analysis / opinion / academic), decomposes claims into elements, maps evidence to those elements, and presents the full landscape — structured, not summarised.

Mission: "We organise; you decide."

Quick Start

Prerequisites
  • Docker & Docker Compose
  • Python 3.11+
  • Node.js 20+
1. Start Infrastructure
docker-compose up -d    # PostgreSQL 16, Redis 7, Qdrant
2. Backend
cd backend
python -m venv venv
source venv/bin/activate    # Windows: venv\Scripts\activate
pip install -r requirements.txt
cp .env.example .env        # Edit with your API keys
alembic upgrade head
uvicorn app.main:app --reload    # http://localhost:8000
3. Frontend
cd web
npm install
npm run dev    # http://localhost:3000

Project Structure

tru8/
├── backend/              # FastAPI backend + evidence pipeline
│   ├── app/
│   │   ├── api/v1/       # REST endpoints (checks, agent, verify, payments)
│   │   ├── core/         # Config, auth, pricing, manifest signing
│   │   ├── models/       # SQLModel schemas (Check, Claim, Evidence, ClaimConsensus)
│   │   ├── pipeline/     # Two-phase evidence pipeline
│   │   ├── services/     # API adapters, search, payments, consensus
│   │   └── utils/        # Classifiers, deduplication, query planning
│   ├── alembic/          # Database migrations
│   ├── tru8_mcp/         # MCP server package for AI agents
│   └── tests/            # 990 unit + integration tests
├── web/                  # Next.js 14 frontend
│   ├── app/              # Pages (dashboard, developers, public reports)
│   ├── components/       # Evidence views, claim map, marketing
│   └── lib/              # API client, tiers, diagnostics
├── shared/               # Shared TypeScript types
├── docker-compose.yml    # PostgreSQL, Redis, Qdrant, MinIO (optional)
└── .claude/CLAUDE.md     # Full engineering context

Pipeline

Two-phase pipeline with user claim selection gate (article mode):

Phase 1:  Ingest → Extract claims → Classify article → Rank claims
          [PAUSE — user selects claims]

Phase 2:  Fact-check lookup → Decompose into elements → Retrieve evidence →
          Score relevance → Classify tier/type → Map evidence to elements →
          Derive orientation → Complete

Post:     Coverage recovery (Stage 5.1), video recommendations, auto-archiving

Evidence flows through web search plus specialist source adapters:

CategorySources
Web searchSerper.dev, Brave Search, SerpAPI (fallback chain)
Fact-checkGoogle Fact-Check API
GovernmentGOV.UK, Hansard, GovInfo, Companies House, ONS
LegalUK Legislation (legislation.gov.uk)
EconomicFRED, Marketaux, World Bank
AcademicSemantic Scholar, OpenAlex, PubMed
ClimateNOAA, WeatherAPI, Open-Meteo, GBIF
ArchivesWikipedia (+ reference mining), Library of Congress, Internet Archive
HealthWHO
SportsTransfermarkt, Football-Data.org
VideoYouTube Data API

Six Evidence Views

ViewQuestion it answers
CartographerWhat's the shape of this conversation?
LibrarianIs this the full set, clearly labelled?
InterpreterDoes this evidence answer this sub-question?
ProjectionistWhat's been said on camera?
ChronologistWhen did the evidence appear?
SeekerWhat don't we know yet?

API & Agent Access

Tru8 exposes a developer API and MCP server for AI agents.

Agent Commerce Gateway
# Quick check (~15s, heuristic classification)
curl -X POST https://api.trueight.com/api/v1/agent/quick \
  -H "X-API-Key: $TRU8_API_KEY" \
  -d '{"claim": "UK inflation is 3.2%"}'

# Smart endpoint with tier fallback (lookup → consensus → quick → full)
curl -X POST https://api.trueight.com/api/v1/agent/check \
  -H "X-API-Key: $TRU8_API_KEY" \
  -d '{"claim": "UK inflation is 3.2%", "max_tier": "full"}'
TierCostTimeWhat you get
Lookup~£0.02instantCached prior analysis
Consensus~£0.03instantCross-user aggregate landscape (k>=3 checks)
Quick~£0.07~15sWeb search + heuristic classification
Full~£0.15~60-90sweb + specialist APIs, LLM classification, element decomposition

Three payment rails: x402 (USDC/SIWE), Skyfire (JWT), prepaid credits.

MCP Server
{
  "mcpServers": {
    "tru8": {
      "command": "python",
      "args": ["-m", "tru8_mcp"],
      "env": { "TRU8_API_KEY": "tru8_sk_..." }
    }
  }
}

Tools: tru8_check (evidence research with tier fallback), tru8_get_result (with analytics), tru8_get_result_raw.

Manifest Verification

Every completed check receives an HMAC-SHA256 signed manifest. Verify the signed fields haven't changed since signing:

curl https://api.trueight.com/verify/{check_id}

Testing

cd backend
pytest tests/ -v                        # Full suite (990 tests)
pytest tests/unit/pipeline/ -v          # Pipeline unit tests
pytest tests/unit/agent/ -v             # Agent/commerce tests
pytest tests/integration/ -v            # Integration tests

Infrastructure

ServicePortPurpose
PostgreSQL 165433Primary database (SQLModel)
Redis 76379Cache + rate limiting
Qdrant6333Vector similarity search
MinIO (optional)9000Local S3 for file uploads

Auth: Clerk (JWT + JWKS) for dashboard, API keys for developer access. Payments: Stripe (4 subscription tiers) + Agent per-use payments.

API Documentation

Environment Variables

Copy backend/.env.example to backend/.env. Key groups:

GroupRequiredVariables
DatabaseYesDATABASE_URL
RedisYesREDIS_URL
AuthYesCLERK_SECRET_KEY, CLERK_JWT_ISSUER
LLMYesGOOGLE_AI_API_KEY, OPENAI_API_KEY
SearchYes (at least one)SERPER_API_KEY, BRAVE_API_KEY, SERP_API_KEY
PaymentsFor subscriptionsSTRIPE_SECRET_KEY, STRIPE_WEBHOOK_SECRET
Government APIsOptionalFRED_API_KEY, NOAA_API_KEY, etc.
Manifest signingOptionalMANIFEST_SIGNING_ENABLED, MANIFEST_SIGNING_KEY

Feature flags (all default to sensible values): ENABLE_API_RETRIEVAL, ENABLE_LLM_RELEVANCE_SCORER, ENABLE_QUERY_PLANNING, etc. See backend/.env.example for the full list.

Licence

Private — All rights reserved.

Frequently asked questions

What is Tru8 Evidence Research?

Source-traced evidence research for AI agents. We organise the evidence; you decide.

How do I install Tru8 Evidence Research?

It runs as a remote server. In Claude Code: claude mcp add --transport http tru8-evidence-research 'https://api.trueight.com/mcp/' --header 'X-API-Key: <Tru8 API key (tru8_sk_...). Create at dashboard > Settings > Developer.>'. For other clients, paste the JSON configuration from this page into their MCP settings.

Does Tru8 Evidence Research need an API key?

Yes. It expects TRU8_API_KEY, X-API-Key, which you set in its configuration.

Is Tru8 Evidence Research open source?

Its code is public on GitHub, but the repository does not declare a standard license.

Who maintains Tru8 Evidence Research?

It is published by SamYatesSmith. AI.tn lists it but does not maintain or endorse it.

From the official MCP Registry. Listing is not an endorsement: review a server's code and permissions before giving it access to your data.