Deep Research Brief

Pose a question and a research desk produces a cited, adversarially verified brief — a researcher scopes, fans out parallel searches, and self-critiques; a verifier checks every claim against its source — for you to approve.

Join the waitlist

What it installs

Agents 2

  • Lead Researcher

    A senior research analyst who scopes the question, runs parallel sourced research, and synthesizes a cited brief.

  • Verifier

    An adversarial fact-checker who opens each citation and rejects unsupported or misattributed claims.

Teams 1

  • Research Desk

    A lead researcher who scopes, fans out, and drafts, paired with a verifier who adversarially checks every claim before anything ships.

Workflows 1

  • Scope, research, synthesize, verify, approve

    Scope the question, fan out parallel sourced research, synthesize and self-critique a cited draft, verify every claim, and hand off for approval.

Skills 4

  • deep-research

    Broad parallel web searches, multi-source validation, confidence tracking, and a cited prose report across 11 research types, with citation discipline and write-as-you-go. Adapted from samber/cc-skills/deep-research.

  • company-research

    Strategic company profiles that extract executive quotes, product strategy, transformation initiatives, and org dynamics from public sources — interview, partnership, and market-entry intelligence, not valuation. Adapted from deanpeters/product-manager-skills/company-research.

  • competitor-analysis

    Map a competitive landscape: identify five direct competitors, profile strengths/weaknesses/pricing, and surface differentiation opportunities and a positioning recommendation. Adapted from phuryn/pm-skills/competitor-analysis.

  • dispatching-parallel-agents

    Delegate independent research axes to focused sub-agents with isolated context, dispatched in one message to run concurrently, then review and integrate — focused scope, clear output, no shared state. Adapted from obra/superpowers/dispatching-parallel-agents.

Requirements

What this template expects to do its job. Task Machine does not verify these — you decide whether your setup is ready.

  • Web search and fetch access for web research — Web search is the core capability of this template. Research breadth, citation depth, and verification all depend on the runtime's web browsing tools; without them the desk halts and says so.
  • Parallel sub-agents (recommended) — Quality and speed improve markedly when the researcher can fan out several focused search sub-agents at once, one per research axis, rather than searching sequentially.

Get started

Install Deep Research Brief and run it with approvals.

Join the waitlist and we will send early access when the first private beta spots open.

Private beta. We invite teams in batches and never share your email.