deep-research

Searches multiple web sources via firecrawl and exa MCPs to produce cited research reports.

Updated Mar 26, 2026
One-click install
npx skills add https://github.com/inuishan/PET --skill deep-research-inuishan
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: deep-research
Source: https://github.com/inuishan/PET/tree/main/.codex/skills/deep-research
Command: npx skills add https://github.com/inuishan/PET --skill deep-research-inuishan

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Answering complex questions requires gathering, reading, and synthesizing information from many web sources, which is slow and error-prone when done manually. This Skill automates multi-source web research and delivers structured reports with inline citations and source attribution. ## Core Features & Use Cases - Multi-Source Search: Runs keyword-varied queries through firecrawl and exa MCP tools, targeting 15-30 unique sources per topic. - Deep Source Reading: Fetches full content of key URLs with firecrawl_scrape or crawling_exa instead of relying on search snippets. - Cited Report Generation: Produces a structured report with executive summary, themed findings, key takeaways, source list, and methodology notes. - Parallel Subagent Research: Splits broad topics across parallel research agents for faster coverage. - Use Case: Ask for a deep dive into the competitive landscape for AI code editors and receive a report with cross-referenced claims, recency-filtered sources, and clearly labeled gaps. ## Quick Start Research the current state of nuclear fusion energy and produce a cited report using firecrawl and exa.

Frequently Asked Questions about deep-research

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I do deep web research with citations using AI?▼

Provide a research topic and the skill breaks it into 3-5 sub-questions, searches each with firecrawl or exa, reads key sources in full, and delivers a structured report with inline citations and a source list.

What MCP tools are required for multi-source web research?▼

At least one of firecrawl (firecrawl_search, firecrawl_scrape, firecrawl_crawl) or exa (web_search_exa, web_search_advanced_exa, crawling_exa) must be configured. Using both together gives the best coverage.

How do I configure firecrawl or exa MCP servers?▼

Add the MCP server configuration to ~/.claude.json or ~/.codex/config.toml with your API credentials. The skill detects whichever tools are available at runtime.

Can research tasks run in parallel across multiple agents?▼

Yes, broad topics can be split across parallel research agents using the Task tool, with each agent handling a subset of sub-questions. The main session then synthesizes their findings into the final report.

What are the limitations of automated web research reports?▼

Reports depend on publicly indexed web content and available search results, so paywalled or very recent information may be missed. Single-source claims are flagged as unverified, and gaps are explicitly acknowledged.