News Evidence Research PlatformTRU8

News Evidence Research Platform

Look behind the headlines.
Form your view.

Tru8 isn't a fact checker. Headlines make claims every day, and the evidence behind them is scattered across dozens of sources. Tru8 gathers that evidence, classifies it by proximity and type, and organises the full landscape — so you don't have to. We organise. You decide.Building an agent? The same evidence research is available as a structured API — read the dev docs.

Submit

Submit an Article or Claim

Paste a news article, headline, claim, or URL you want to research.

Research

Evidence Gathered

Tru8 searches government data, news archives, academic papers, and official records. Each piece of evidence is classified by proximity and type.

Explore

Explore the Landscape

Six views of the same evidence — in the browser, or as JSON via the API. See the shape, browse the collection, focus on specifics, watch the conversation, trace the timeline, or surface what’s missing.

Your Research Team

Six ways to explore

The Cartographer

“What’s the shape of the conversation?”

See where sources agree, where they diverge, and which are just echoing the same original.

The Librarian

“Show me the full set, clearly labelled.”

Every source classified by proximity and type. Filter, sort, browse. Nothing hidden.

The Correspondent

“Who’s in the room?”

See which domains contributed evidence, how concentrated or diverse they are, and where single-source coverage may need attention.

The Projectionist

“What’s being said about this on camera?”

Relevant video context from YouTube, classified the same way as text sources.

The Chronologist

“When did each piece of evidence appear?”

A timeline of every source, ordered by publication date. See how the conversation developed and where the reporting clusters.

The Seeker

“What don’t we know yet?”

Every evidence gap, surfaced clearly. Specify what data would fill each one, then trigger a targeted re-search.

Module — See It Work

Four views.
One landscape.

The same evidence research, surfaced four ways. Click any panel to open it full-size with zoom.

01 / 04

Librarian

Every source classified.

Tier × type heatmap and ledger. Filter, sort, browse. Receipts for everything excluded — no hidden curation.

02 / 04

Cartographer

The shape of the conversation.

A citation cascade. See where sources agree, where they diverge, and which are just echoing the same original.

03 / 04

Seeker

What we don’t know yet.

Every evidence gap, surfaced clearly. Specify what data would fill each one, then trigger a targeted re-search.

04 / 04

Chronologist

When did each piece of evidence appear?

A timeline of every source, ordered by publication date. See how the conversation developed and where the reporting clusters.

Module — For Builders

One submission,
structured for agents.

The same evidence research, returned as a stable JSON contract. Per-claim element decomposition, computed landscape metrics, and a signed manifest your callers can verify.

POST /agent/quick — Response200 OK

{
  "id": "chk_8f3a...",
  "status": "complete",
  "claims": [
    {
      "id": "clm_01",
      "text": "Global average temperature rose 1.1°C since pre-industrial times",
      "claimMap": {
        "normalisedClaim": "...",
        "claimType": "scientific",
        "elements": [
          {
            "elementId": "el_01",
            "description": "1.1°C rise figure",
            "state": "supported",
            "evidenceRefs": [
              { "evidenceId": "ev_a1", "relationship": "supports" },
              { "evidenceId": "ev_b2", "relationship": "supports" }
            ]
          }
        ],
        "orientation": "Evidence converges on the 1.1°C figure across primary and academic sources."
      }
    }
  ],
  "_meta": {
    "executedTier": "quick",
    "chargedPence": 7,
    "landscape": {
      "elementCount": 4,
      "elementStates": { "supported": 3, "unresolved": 1 },
      "evidenceDensity": 24,
      "sourcesConsidered": 24,
      "sourceDiversity": {
        "tierSpread": { "primary": 6, "reporting": 12, "commentary": 6 },
        "uniqueDomains": 18,
        "typeCoverage": 5
      },
      "freshness": { "freshestDaysAgo": 3, "dateSpanDays": 412 },
      "gaps": [{ "reason": "no_academic_sources" }]
    },
    "limitations": ["heuristic_classification", "single_query_per_element"]
  },
  "_manifest": {
    "checkId": "chk_8f3a...",
    "landscapeHash": "9c14...",
    "signature": "hmac-sha256:...",
    "verifyUrl": "/verify/chk_8f3a..."
  }
}

What’s in the response

claimMap
Per-claim decomposition. Elements, evidence refs with relationship, mechanical orientation line.
_meta.landscape
Computed metrics — element states, tier spread, unique domains, freshness, identified gaps.
_manifest
Signed payload + verify URL. Your downstream caller can prove what you sent them.

Curl — one call

curl -X POST https://api.trueight.com/agent/quick \
  -H "Authorization: Bearer $TRU8_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{ "claim": "Global average temperature rose 1.1°C since pre-industrial times" }'

Read the docs

Per-call pricing from £0.02
Signed manifests, MCP server, webhooks

Plans

Choose your plan

Free Trial

Try Tru8 with 3 free checks

£0/mo

3 evidence checks
All source types
All six views

Starter

For regular research

£7/mo

40 checks per month
All source types
All six views
Export reports

Professional

High-volume evidence research

£29/mo

200 checks per month
Full API & MCP access
Priority processing
Export reports

Enterprise

Custom volume, SLA, and support

Custom

Custom check volume
Dedicated support
SLA guarantee
Custom integrations

Look behind the headlines.Form your view.