deep-research

Featured

Deep research harness — fan-out web searches, fetch sources, adversarially verify claims, synthesize a cited report.

AI & Automation 68,373 stars 11111 forks Updated 3 days ago CC0-1.0

Install

View on GitHub

Quality Score: 91/100

Stars 20%
100
Recency 20%
100
Frontmatter 20%
70
Documentation 15%
48
Issue Health 10%
50
License 10%
100
Description 5%
100

Skill Content

Run the "deep-research" workflow. Deep research harness — fan-out web searches, fetch sources, adversarially verify claims, synthesize a cited report. When the user wants a deep, multi-source, fact-checked research report on any topic. BEFORE invoking, check if the question is specific enough to research directly — if underspecified (e.g., "what car to buy" without budget/use-case/region), ask 2-3 clarifying questions to narrow scope. Then pass the refined question as args, weaving the answers in. Phases: - Scope: Decompose question (from args) into 5 search angles - Search: 5 parallel WebSearch agents, one per angle - Fetch: URL-dedup, fetch top 15 sources, extract falsifiable claims - Verify: 3-vote adversarial verification per claim (need 2/3 refutes to kill) - Synthesize: Merge semantic dupes, rank by confidence, cite sources Invoke: Workflow({ name: "deep-research" }) ## Workflow Script [scripts/workflow-script.js](scripts/workflow-script.js)

Details

Author
asgeirtj
Repository
asgeirtj/system_prompts_leaks
Created
1 years ago
Last Updated
3 days ago
Language
JavaScript
License
CC0-1.0

Integrates with

Similar Skills

Semantically similar based on skill content — not just same category

AI & Automation Listed

deep-research

Use when executing thorough multi-step web research, paper synthesis, or market analysis

7 Updated yesterday
imMamdouhaboammar
AI & Automation Listed

deep-research

Deep research harness — fan-out web searches, fetch sources, adversarially verify claims, synthesize a cited report. When the user wants a deep, multi-source, fact-checked research report on any topic. BEFORE invoking, check if the question is specific enough to research directly — if underspecified (e.g., "what car to buy" without budget/use-case/region), ask 2-3 clarifying questions to narrow scope. Then pass the refined question as args, weaving the answers in.

8 Updated today
ek33450505
Data & Documents Listed

deep-research

Run a deep, multi-source, fact-checked research investigation that ends in a cited report, not a single web lookup. Use when the user asks to "research X", "do a deep dive on", "find out everything about", "investigate", "compare the options for", "what's the state of the art on", or "give me a report with sources" — or poses a broad, open-ended question that one search cannot settle and that needs several sources cross-checked against each other. Runs the pipeline scope -> decompose into angles -> fan out searches -> dedupe and fetch sources -> extract falsifiable claims -> adversarially verify each load-bearing claim -> synthesize a confidence-ranked, per-claim-cited report that keeps negative findings. Prefer it over an ad-hoc search whenever being wrong is costly or the answer must be defensible.

0 Updated yesterday
hsb3