← ALL AI EMPLOYEES

Intelligence · Tools · MODULE RESEARCH

Live Web Research

Live web research, callable from any module — no stale data

Module RESEARCH is the live-search backbone. When Module N needs to research a new lead, when Module COUNCIL needs to fact-check a plan, when your team needs current information — RESEARCH runs the query, aggregates results, and returns structured findings. Included with Module COUNCIL; standalone on request.

Live

SerpAPI · Bing · DuckDuckGo

Structured

JSON findings, not raw HTML

Cited

Every fact links to its source

WHAT THIS FIXES

The exact problems Live Web Research is built around.

  • LLMs trained months ago can't answer "who did our competitor announce as a partner last week".
  • Your team manually searches 5 engines + 10 sites for one piece of competitive intel.
  • Hallucinated facts in proposals burn trust with prospects — your sales team can't cite an LLM.
  • Module N proposals benefit from live context but raw search results need triangulation.

HOW IT WORKS

The pipeline, end to end.

  1. 1

    Query intent

    Module N (or any module) issues a research intent: "Find recent news about <company>, <year>". Intent typed at the schema level.

  2. 2

    Multi-engine fetch

    Module RESEARCH queries SerpAPI (Google), Bing, DuckDuckGo. Top 10 results from each, de-duplicated.

  3. 3

    Fact extraction

    Each result page summarised by Claude Sonnet 4.6. Named entities + dated facts extracted into structured JSON.

  4. 4

    Triangulation

    Facts cross-referenced across sources. Confidence score: high (3+ sources agree), medium (2), low (1, single-source).

  5. 5

    Structured return

    JSON object: { facts: [...], confidence: ..., citations: [...] } returned to the calling module within 12-30s.

WHAT'S INCLUDED

Every capability shipped, named.

  • Multi-engine fetch (SerpAPI Google, Bing, DuckDuckGo)
  • Structured JSON output — not raw HTML
  • Per-fact confidence score (cross-source triangulation)
  • Named entity extraction (people, orgs, dates, places)
  • Per-tenant rate limit and audit log
  • Powers Module N (proposal research) + Module COUNCIL (planning)
  • Direct callable via API for custom use
  • Per-source domain filtering (block junk sources)
  • PDPA-compliant — no PII extracted from search results
  • Cached results (24h TTL) for efficiency

PRICING

Plans that scale with use, not seats.

All prices in MYR, exclusive of SST. 14-day refund on monthly, pro-rated refund on annual. PDPA-aligned. Cancel anytime.

MOST POPULAR

Bundled

Included with Council

Used internally by other modules

See Module COUNCIL

Standalone

RM 599/mo

1,000 queries/mo · API access · per-domain filtering

Talk to sales

WHAT YOU RECEIVE

The artefacts that land in your hands.

Structured JSON

{ facts: [{ fact, confidence, citations }], named_entities: { ... } } — drop into your proposal or report.

Citation list

Every fact backed by 1-5 source URLs with publication date. Auditable.

Audit log

Every query logged: who asked, what was asked, what came back. Compliance-friendly.

FREQUENTLY ASKED

The questions buyers ask first.

  • Why not just call SerpAPI directly?

    You could. Module RESEARCH adds: multi-engine de-dup, fact extraction, triangulation, structured JSON, audit, and tenant rate limits. ~80% of the work after SerpAPI.
  • Latency?

    p50 ~14s · p95 ~28s. Most time spent in fact extraction (Claude pass). Cached results return in <500ms.
  • Hallucinations on extracted facts?

    Cross-source triangulation catches most. Single-source facts marked confidence=low and surfaced as such. Module N's proposal generator filters facts below confidence=medium by default.
  • Per-source filtering?

    Yes — Enterprise can block domains (e.g. block Wikipedia, block competitor blog) or allow-list trusted sources only.
  • Use it directly from my own code?

    Yes — Standalone tier exposes a REST API + MCP server. Drop into your own LLM agent stack.
  • Anti-scraping ToS?

    We use SerpAPI which has commercial Google Search API rights. No raw scraping of Google. ToS-compliant.

Try Live Web Research on your real business.

14-day trial. No credit card. PDPA-aligned. Built for Malaysian SMEs by a Malaysian team in KL. Reginald and ONSET team reply within one business hour, Mon–Fri 9am–6pm MYT.

  • PDPA-aligned
  • Cancel anytime
  • 14-day refund
  • 1 business hour reply