firecrawl-research-index

Find research papers answering a query using semantic search, citation expansion, and in-body verification.

1|Updated Aug 24, 2026
One-click install
npx skills add https://github.com/CliffVale/opencode-free-setup --skill firecrawl-research-index-cliffvale
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: firecrawl-research-index
Source: https://github.com/CliffVale/opencode-free-setup/tree/main/skills/firecrawl-research-index
Command: npx skills add https://github.com/CliffVale/opencode-free-setup --skill firecrawl-research-index-cliffvale

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Finding the right research papers for a question is hard: a single keyword search misses sibling methods, citing work, and papers whose abstracts don't state the property you care about. This Skill turns a research query into a complete, verified set of relevant papers instead of one lucky hit. ## Core Features & Use Cases - Semantic paper search: HyDE-based search over paper abstracts via Firecrawl Research, with guidance on reframing queries when results are thin. - Structural and semantic expansion: Grow one good hit into a full set using similar-paper, citers, and references modes ranked to your intent. - In-body verification: Read specific passages inside a paper to confirm a method was actually used, a score was actually reported, or an affiliation matches. - Use Case: Ask "what are alternatives to Adam for optimizer design" and get an expanded family of relevant papers, each checked for relevance, rather than a single top match. ## Quick Start Use the firecrawl-research-index skill to find all papers that propose training-free methods for detecting AI-generated text, including closely related follow-up work.

Frequently Asked Questions about firecrawl-research-index

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

FAQPage Schema
How do I find research papers on a specific topic?▼

Start with firecrawl research search-papers, which runs semantic HyDE search over paper abstracts. If results look thin or repetitive, re-run with a different framing such as a sibling domain, rival method, or benchmark name.

How to find papers related to a paper I already know?▼

Use firecrawl research related-papers with the known paper's ID as a seed. Choose mode=similar for niche siblings, mode=citers for work that builds on it, or mode=references for what it builds on and compares against.

How do I verify a paper actually reports a specific result?▼

Use firecrawl research read-paper with the paper ID and a targeted question to pull in-body passages. This confirms whether a method was actually used or a score actually reported, rather than relying on the abstract alone.

Can I find the best-performing papers on a benchmark?▼

Yes, but rankings live on leaderboards rather than in abstracts. Use firecrawl search or scrape to find the benchmark leaderboard, read off the top entries, then map each back to its paper with search-papers.

When should I not use semantic paper search alone?▼

Single searches fail for enumeration queries, exhibiting-property queries, and superlative questions. These need citation expansion from multiple seed papers or web leaderboard lookups, since the relevant information is not in any one abstract.