Skip to content
Get packs

Deep Research Orchestrator

Runs multi-source, cited, fact-checked research by splitting work across sub-agents and writing an HTML report.

Source Download skill
Repository
xerrors/Yuxi
Stars
7,576
Licence
MIT
Files
1 file

What it does

When installed, the agent stops answering deep questions from memory and instead follows a research workflow: clarify the scope, split the question into sub-questions, hand each to a parallel research helper, have key claims checked adversarially, then write one structured report with source citations. By default the report is saved as an HTML file in the project folder and presented as a deliverable rather than pasted into chat.

Great for

  • Researchers preparing literature reviews or topic surveys that need traceable sources
  • Analysts doing competitor, industry, or technology-selection research
  • Anyone who needs claims checked and gaps in evidence clearly flagged

Skip it if

  • Quick factual questions, since the multi-session workflow is slow and token-heavy
  • Tasks where the sources must be verified by a human anyway and the agent's citations are not enough to rely on, because the skill relies on the model's own checking
  • Environments without a web_search tool or sub-agent session support, because the workflow depends on them

Try asking

  • “Do a deep research report on the current state of solid-state battery technology, with sources.”
  • “Compare three open-source vector databases for a technology selection report, with citations and fact-checking.”
  • “Survey recent papers on retrieval-augmented generation and write a structured literature review as an HTML file.”
Step-by-step guide: how to use Deep Research OrchestratorInstall, first run, examples, tips and FAQs

Install in 3 steps

  1. 1

    Download the skill

    The ZIP contains the 1 original files, the MIT licence and an attribution note.

    deep-research.zip
  2. 2

    Add it to your agent

    Claude Code (all projects)

    mkdir -p ~/.claude/skills && unzip deep-research.zip -d ~/.claude/skills/

    Claude Code / Agent SDK (this project only)

    mkdir -p .claude/skills && unzip deep-research.zip -d .claude/skills/

    Claude apps (web/desktop): Settings → Capabilities → Skills → Upload skill, and choose the ZIP.

  3. 3

    Ask for the task

    No special command needed — the agent reads the skill's description and loads it when your request matches. Try one of the prompts above.

Get better results

  • Give a clear goal, audience, and time range up front so the clarification step is short.
  • Ask for a specific output format if you do not want the default HTML file.
  • Check the cited sources yourself for any high-stakes numbers, since the verification step is performed by the same model.

What's inside

  • SKILL.md

Redistributed unmodified from xerrors/Yuxi under the MIT licence. The assessment was drafted with AI from the skill's files on 2026-10-29. Skills can instruct your agent to run code — read SKILL.md and any scripts before installing.