research_skill
- Python
6
GitHub Stars
1
Bundled Files
2 months ago
Catalog Refreshed
4 months ago
First Indexed
Readme & install
Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.
Installation
Preview and clipboard use veilstrat where the catalogue uses aiagentskills.
npx veilstrat add skill acking-you/myclaude-skills --skill research- SKILL.md7.9 KB
Overview
This skill performs structured web research with multi-source validation and delivers citation-backed reports. It prioritizes authoritative, recent sources, highlights potential bias, and produces clear, actionable summaries tailored to your required depth. Reports include clickable citations and link validation to ensure verifiable claims.
How this skill works
I scope the question with you, select keywords and domain constraints, then run parallel searches across factual, comparative, technical, practical, and recent angles. Collected sources are diversified (official docs, independent analysis, community threads, academia), cross-referenced for at least two confirmations per claim, and rated for authority, recency, specificity, and independence. Final output is a structured report with validated links, conflict notes, and recommendations.
When to use it
- Quick factual checks or short comparisons (3–5 sources).
- Technical feature comparisons or vendor analysis (standard 8–12 sources).
- Architecture decisions or comprehensive studies (deep 15+ sources).
- When you need verifiable, citation-rich summaries for reports or decision-making.
- When you want vendor bias, version awareness, and link validation included.
Best practices
- Specify scope, depth (Quick/Standard/Deep), and any domain/year constraints up front.
- Request specific priorities: performance, cost, architecture, or use cases to focus searches.
- Prefer official documentation first; use third-party sources to fill gaps and triangulate claims.
- Insist on link validation for any production use; I can check URLs or let you verify interactive links.
- Flag any vendor-produced benchmarks so I can annotate bias and seek independent verification.
Example use cases
- Produce a 1–2 paragraph quick brief comparing two SaaS features with 3–5 citations.
- Deliver a standard report comparing search engine architectures with 8–12 verified sources and a recommendation section.
- Compile a deep research dossier on platform migration with 15+ sources, cross-references, and trade-off analysis.
- Validate claims in a vendor whitepaper by cross-referencing official docs, independent tests, and community issues.
- Create a citation-backed FAQ or internal decision memo with clickable references and link checks.
FAQ
I prioritize official documentation and announcements, then reputable technical analyses and community validation. Each source is scored by authority, recency, specificity, and independence.
Can you verify every link for me?
Yes. I can batch-validate links using page fetch tools and replace broken URLs with alternatives, or provide the links for you to click and confirm.