New: 97,000+ skills indexed from GitHub repos with 100+ stars · every skill safety-checkedBrowse top skillsTop skills
Esc to closeBrowse all skills
Submit a skillBrowse skills
Submit a skillBrowse skills

Agent skill

Deep Research

by XiaomiMiMo in XiaomiMiMo/MiMo-Code

Runs a multi-source investigation with parallel sub-agents and built-in web tools, then writes one cited report. Meant for open-ended topics, not quick lookups.

MITAuto-check passedResearch & Science

Install Deep Research

skills CLI
$ npx skills add XiaomiMiMo/MiMo-Code --skill deep-research -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install XiaomiMiMo/MiMo-Code deep-research --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/XiaomiMiMo/MiMo-Code.git skills-src && mkdir -p .claude/skills && cp -r skills-src/packages/cli/src/skill/builtin/.bundle/deep-research .claude/skills/deep-research && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
deep-research
GitHub stars
14k
Token cost
~1.2k tokens
SKILL.md length
555 words
Files
4
Skills in repo
22
Repo updated
First seen
Licence
MIT

At a glance

Runs a multi-source investigation with parallel sub-agents and built-in web tools, then writes one cited report. Meant for open-ended topics, not quick lookups.

  • Works in 6 steps: Always first → Scope → Plan → …
  • Investigating a topic across many sources and getting a cited report
  • SKILL.md covers Step 0 — Always first, Workspace, Phase 1 — Scope and Phase 2 — Plan, plus 3 more sections
  • Instructions only: no scripts, shell commands, URLs or credentials in SKILL.md

What it does

This skill orchestrates several research sub-agents in parallel and then has a single writer produce one coherent report, so sections never come from competing agents. It starts by checking today's date with a shell command and triaging the request: a question that one or two searches can answer is sent back to plain WebSearch, while large enumeration tasks use one sub-agent per batch of items. Three depth modes set the budget: quick uses two to three sub-agents with no follow-up round, standard three to five with one, and deep five to eight with two.

All state is written to disk in a per-topic research folder holding a brief and one findings file per sub-agent, so work survives context compaction and can be resumed. The scoping phase allows at most one round of clarifying questions and records assumptions in the brief. Planning splits the brief into three to eight independent angles, which can be shown to you for confirmation in deep mode. The skill uses only built-in search and fetch tools plus free APIs that need no keys, and academic literature surveys go to a different skill.

When your agent uses it

  • Investigating a topic across many sources and getting a cited report
  • Comparing a long list of items on several fields with parallel agents
  • Verifying a contested claim from several independent angles
  • Resuming an unfinished research run from its saved brief and findings

Example prompts

  • “Do a deep research run on the state of open-source vector databases and write a cited report.”
  • “Compare twenty JavaScript charting libraries on license, bundle size and maintenance.”
  • “Run a quick investigation into whether the new EU data rule applies to small shops.”

Requirements

  • WebSearch and WebFetch tools
  • The ability to run parallel sub-agents

Workflow steps

6 steps, taken from the step headings in SKILL.md.

  1. Always first
  2. Scope
  3. Plan
  4. Parallel research
  5. Reflect (gap check)
  6. Write (single-point)

What it can do on your machine

Read from SKILL.md and the folder at commit 6babeb0. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    No scripts in the folder and no shell commands in SKILL.md.

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Deep Research loads about 1.2k tokens when it runs. Until then it costs about 100 tokens; SKILL.md has 555 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~100
When it runs · the whole SKILL.md, loaded when a task matches
~1.2k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from XiaomiMiMo/MiMo-Code at commit 6babeb0, republished under its MIT licence (© XiaomiMiMo). 555 words, ~1,184 tokens.

Download SKILL.mdSave it as .claude/skills/deep-research/SKILL.md (or your agent's skills folder). This skill also uses 3 other files; get the full folder from GitHub.
name
deep-research
description
Deep research on any topic using parallel sub-agents and built-in tools only (WebSearch/WebFetch + free APIs, no keys). Use when the user asks for a thorough multi-source investigation with a cited report — "深度调研X"、"deep research"、"帮我全面研究一下"、"多方求证"、"写一份调研报告". NOT for simple lookups (single WebSearch suffices) and NOT for academic literature surveys (use auto-research skill instead).

Deep Research

Orchestrate parallel research sub-agents, then write one coherent cited report. Research is parallel; writing is single-point — never let multiple agents write report sections.

Step 0 — Always first

  1. Run date +%Y-%m-%d via Bash. Never assume the current year from training data.
  2. Triage:
    • Answerable with 1-2 searches? → STOP, just use WebSearch directly. Do not use this skill.
    • Enumeration task (N items × M fields, e.g. "compare 20 frameworks")? → still this skill, but use table-oriented decomposition (one sub-agent per item batch).
    • Open-ended investigation? → continue below.
  3. Pick depth (default standard; user can override with words like "quick"/"exhaustive"):
ModeSub-agents (round 1)Max follow-up roundsSources target
quick2-308+
standard3-5115+
deep5-8225+

These are hard budgets. Reflection (Phase 4) can spend them but never exceed them.

Workspace

All state lives on disk at ./research/<slug>/ — never only in context (survives compaction):

research/<slug>/
├── brief.md         # research brief — the single contract for all phases
├── findings/        # F1.md, F2.md ... one per sub-agent, structured evidence
└── REPORT.md        # final deliverable

On resume: re-read brief.md + list findings/, skip completed angles, continue.

Phase 1 — Scope

Ask at most one round of clarifying questions (AskUserQuestion), only if genuinely ambiguous: audience, time frame, region, decision at stake. If the user said "just run it" or intent is clear, skip asking and write assumptions into the brief instead.

Then write brief.md: refined question, scope boundaries (in/out), assumptions, depth mode, today's date. This brief — not the raw conversation — is what every later phase measures against.

Phase 2 — Plan

Decompose the brief into 3-8 independent research angles. Pull from these lenses as applicable: core facts/definitions · recent developments (last 12 months) · quantitative data/benchmarks · counter-arguments & failure cases · practitioner experience (forums, issues) · academic work · key players/alternatives.

List angles in brief.md under ## Angles. For deep mode or contested topics, show the angle list to the user for a quick confirm before spending budget.

Show full SKILL.md (267 more words)Show less

Phase 3 — Parallel research

Spawn one sub-agent per angle in a single message (parallel). Build each prompt from the locked template in reference/subagent-prompt.md — reproduce it verbatim, replacing only the {variables}. Each sub-agent:

  • researches ONE angle only, using WebSearch/WebFetch and the free endpoints in reference/sources.md
  • writes structured findings to findings/F<n>.md (claim / quote / URL / date / confidence per item)
  • returns only a 3-5 line summary to you — raw page content must never enter your context

If a sub-agent fails or returns thin results, note it and move on; do not block other angles.

Phase 4 — Reflect (gap check)

Read all findings/*.md. Against brief.md, ask: which parts of the brief have no evidence? Which major claims rest on a single source? Where do sources conflict?

  • Gaps found AND follow-up budget remains → spawn targeted sub-agents with delta-queries (same template, narrower angle). Repeat once per remaining round.
  • No budget left or coverage sufficient → proceed. Record unresolved gaps; they go in the report's "Open questions".

Phase 5 — Write (single-point)

You alone write REPORT.md in one pass, following reference/report.md. Core rules:

  • Every non-obvious claim carries an inline citation [n] mapping to a Sources section; citation URLs come only from findings files — never from memory.
  • Where sources conflict, present both sides with dates; prefer newer + primary sources.
  • Mark single-source claims with [single source], speculation with [speculative].
  • End with: Open questions · Sources (numbered, with access date).

For deep mode, before finalizing do one critique pass: reread the report as a skeptical reviewer (unsupported claims? stale data? missing counter-view?) and fix in place.

Finally, give the user a 5-10 line summary of key conclusions and the report path.

© XiaomiMiMo, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 3 other files in packages/cli/src/skill/builtin/.bundle/deep-research of XiaomiMiMo/MiMo-Code.

  • SKILL.md
  • reference/report.md
  • reference/sources.md
  • reference/subagent-prompt.md

Open the folder on GitHubat commit 6babeb0

Compare with similar skills

Deep Research next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Deep Research compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Deep Research this skillXiaomiMiMo/MiMo-Code14k—~1.2kAutomated safety check: PassMIT
ULW Deep Researchcode-yeongyu/oh-my-openagent70k—~14kAutomated safety check: PassCustom licence
Deep Research Teammalob/nix-config4631 repos~5.8kAutomated safety check: PassMIT
Advanced Swarm Orchestrationruvnet/agentic-flow8165 repos~5.9kAutomated safety check: PassNone
Web ResearchJuncai22/spring-ai-agent-learning1233 repos~1.1kAutomated safety check: PassApache-2.0
Deep Research312362115/claude107—~6.6kAutomated safety check: PassMIT

Similar skills

  • ULW Deep Research

    code-yeongyu/oh-my-openagent

    Runs an exhaustive, team-based research session that stands up cooperating agents, debates findings and delivers a report where every claim has a citation or proof.

    70k GitHub stars~14k tokensUpdated today
    Research & ScienceAuto-check passed

  • Deep Research Team

    malob/nix-config

    Coordinates a team of researcher agents across several rounds, with a lead who triages findings, assigns follow-ups and cross-checks key claims.

    463 GitHub starsUsed in 1 repo~5.8k tokens
    Research & ScienceAuto-check passed

  • Advanced Swarm Orchestration

    ruvnet/agentic-flow

    Patterns for running multi-agent swarms on research, development and testing work, with four topologies and a four-phase research swarm built on claude-flow MCP tools.

    816 GitHub starsUsed in 5 repos~5.9k tokens
    Agent WorkflowsAuto-check passed

  • Web Research

    Juncai22/spring-ai-agent-learning

    A skill your agent uses for requests related to web research; it provides a structured approach to conducting comprehensive web research

    123 GitHub starsUsed in 3 repos~1.1k tokens
    Research & ScienceAuto-check passed

  • Deep Research

    312362115/claude

    深度调研技能：对任意命题进行系统性调研并输出专业研究报告. An agent skill from 312362115/claude.

    107 GitHub stars~6.6k tokensUpdated 4 mo ago
    Research & ScienceAuto-check passed

  • Deep Dive

    rohitg00/awesome-claude-code-toolkit

    Claude-native deep research using DAG-based query planning, parallel subagent execution, and gap-driven iteration.

    2.7k GitHub stars~1.5k tokensUpdated 4 mo ago
    Research & ScienceAuto-check: warnings

More from XiaomiMiMo/MiMo-Code

All 22 skills in this repo

  • Paper Research on arXiv

    XiaomiMiMo/MiMo-Code

    Searches arXiv, fetches metadata, generates BibTeX, downloads PDFs and finds citations and related papers using a bundled Python script.

    14k GitHub starsUsed in 1 repo~1.5k tokens
    Auto-check passed

  • Agent Skill Creator

    XiaomiMiMo/MiMo-Code

    Interactive guide for creating, reviewing and fixing agent skills (SKILL.md folders), covering structure, frontmatter rules, trigger phrases and validation before sharing.

    14k GitHub stars~1.9k tokensUpdated 4 days ago
    Auto-check passed

  • DOCX Toolkit

    XiaomiMiMo/MiMo-Code

    Produces, edits and reads Microsoft Word files through python-docx and lxml, with a decision table for picking the lightest workflow for a given task.

    14k GitHub stars~2.4k tokensUpdated 4 days ago
    Auto-check passed

  • Drive MiMo Code

    XiaomiMiMo/MiMo-Code

    Lets one MiMoCode process drive another, headless with JSON events or interactively through tmux, to test behavior and visual regressions with parseable evidence.

    14k GitHub stars~3.9k tokensUpdated 4 days ago
    Auto-check passed

  • PDF Toolkit

    XiaomiMiMo/MiMo-Code

    Reads, transforms, composes and fills PDFs with Python scripts for extraction, merging, watermarking, encryption, OCR and form filling.

    14k GitHub stars~1.7k tokensUpdated 4 days ago
    Auto-check passed

  • XLSX Spreadsheet Toolkit

    XiaomiMiMo/MiMo-Code

    Builds, edits, cleans, recalculates and reads Excel workbooks and CSV files with openpyxl and pandas, plus LibreOffice for recalculation and PDF export.

    14k GitHub stars~2.9k tokensUpdated 4 days ago
    Auto-check passed

Categories

Questions about Deep Research

What does Deep Research do?

Runs a multi-source investigation with parallel sub-agents and built-in web tools, then writes one cited report. Meant for open-ended topics, not quick lookups. This skill orchestrates several research sub-agents in parallel and then has a single writer produce one coherent report, so sections never come from competing agents. It starts by checking today's date with a shell command and triaging the request: a question that one or two searches can answer is sent back to plain WebSearch, while large enumeration tasks use one sub-agent per batch of items.

When should I use Deep Research?

Deep Research fits situations like: investigating a topic across many sources and getting a cited report; comparing a long list of items on several fields with parallel agents; verifying a contested claim from several independent angles; resuming an unfinished research run from its saved brief and findings.

How do I install Deep Research in Claude Code?

Run `npx skills add XiaomiMiMo/MiMo-Code --skill deep-research -a claude-code`. Or copy the skill folder (packages/cli/src/skill/builtin/.bundle/deep-research in XiaomiMiMo/MiMo-Code) into .claude/skills/deep-research in your project. Claude Code loads it when a task matches its description.

How do I install Deep Research in Codex?

Run `npx skills add XiaomiMiMo/MiMo-Code --skill deep-research -a codex`. Or copy the skill folder (packages/cli/src/skill/builtin/.bundle/deep-research in XiaomiMiMo/MiMo-Code) into .agents/skills/deep-research in your project. Codex loads it when a task matches its description.

Can I use Deep Research in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add XiaomiMiMo/MiMo-Code --skill deep-research -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/deep-research, .gemini/skills/deep-research, .github/skills/deep-research and .opencode/skills/deep-research in your project.

What does Deep Research need to run?

SKILL.md names no scripts, command-line tools or credentials: Deep Research is instructions for the agent only. Our summary lists: WebSearch and WebFetch tools; The ability to run parallel sub-agents.

Does Deep Research access the network?

SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.

Is Deep Research safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Deep Research use?

Deep Research is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Deep Research use?

About 1.2k tokens (SKILL.md is roughly 4.7k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to Deep Research?

Skills that share tags, products or a category with Deep Research: ULW Deep Research (code-yeongyu/oh-my-openagent, 70k stars), Deep Research Team (malob/nix-config, 463 stars), Advanced Swarm Orchestration (ruvnet/agentic-flow, 816 stars) and Web Research (Juncai22/spring-ai-agent-learning, 123 stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Deep Research?

XiaomiMiMo (a GitHub organization) maintains it in XiaomiMiMo/MiMo-Code, which has 13,601 GitHub stars. The repository holds 22 skills in this directory. The repository was last updated on October 3, 2026.

Source: XiaomiMiMo/MiMo-Code on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.