# Research

> Conduct comprehensive deep research on a topic using 7-step algorithm Arguments: <topic> [--type <competitive|market|technical|person|topic|news>] [--depth <quick|standard|deep>].

- Skill: `agents-store/research` (Agent Skill)
- Install (CLI): `npx get-skill add agents-store/research`
- Raw SKILL.md: https://api.skillmd.com/api/skills/agents-store/research/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Research & Search
- Author: agents-store (https://skillmd.com/u/agents-store)
- Updated: 2026-11-16
- Page: https://skillmd.com/skills/agents-store/research

---


# Deep Research

Execute the full 7-step research algorithm from the `deep-research` skill: read `${CLAUDE_PLUGIN_ROOT}/skills/deep-research/SKILL.md` first and follow it. See CONNECTORS.md for provider mapping and fallback chains.

## Steps

1. **CLASSIFY** — determine research type from topic (or use --type flag)
2. **PLAN** — generate 3-7 search queries from different angles (a model step, no tool)
3. **SEARCH** — ~~batch_search / ~~search with fallback; ~~answer for facts; ~~academic_search / ~~code_search by type; one ~~deep_agent pass when `--depth deep`
4. **READ** — ~~batch_scrape / ~~scrape top-5 pages with fallback; pass the topic as `question` to read only the relevant passages
5. **EXTRACT** — key facts, data, quotes with source URLs
6. **SYNTHESIZE** — deduplicate, cross-check, assess confidence
7. **REPORT** — structured report using appropriate template + Methodology

For full algorithm details, query patterns per type, and quality checklist — see the `deep-research` skill.

## Example Usage
```
/research AI code assistants market 2026
/research Notion vs Linear vs Asana --type competitive
/research RAG pipeline architecture --type technical --depth deep
/research latest AI regulation news --type news
```
