Deep Research Reports
Research skill, available on Zeplik
Deep Research Reports is a ready-to-run research skill on Zeplik. Plan, search, open the real pages and synthesize a long multi-step cited report -- market analysis, competitive landscaping, literature review, due diligence, comparisons. Ask in plain language and Zeplik applies the skill's method for you inside the conversation, on whichever AI model you prefer. It returns a structured document you can keep and reuse: Document artifact -- structured written deliverable with headed sections and a TL;DR (see artifact-templates/document.md).
The Deep Research Reports skill loads automatically when your request matches it, or you can invoke it directly by typing /deep-research in any chat. It works with attachments, connectors, and any model that supports the task, so you get the same expert method every time without setting anything up.
What the Deep Research Reports skill can do
- Plan and execute autonomous multi-step research across many sources
- Synthesize findings into a cited, structured markdown report
- Stream or poll progress on long-running research tasks
- Continue prior research sessions to elaborate on specific points
Try these prompts on Zeplik
Pick a prompt to open it in the Zeplik app. If you are not signed in yet, your prompt is waiting for you the moment you do.
How the Deep Research Reports skill works
Deep Research
Plan, search, read the real pages, and synthesize into a comprehensive cited report.
When to Use This Skill
- Market analysis and competitive landscaping
- Technical literature reviews
- Due diligence
- Historical research and timelines
- Comparative analysis (frameworks, products, technologies)
Not for a single fact look-up. Use the research skill, or plain web_search, for that.
The tools you actually have
web_search and read_url, both already on the wire for this turn. Nothing else
is needed and nothing else is installed -- there is no research CLI, no external
research API, and no separate research service to call. The depth comes from how
you use these two, not from a tool you do not have.
web_search does not return research. It returns at most eight results, each
one a title and a single line of snippet text. A report written from snippets is a
report written from a search engine results page. It will read plausibly and it
will be shallow, and the reader cannot tell the difference until they check.
read_url is where the research is. It returns the real page text. Every
substantive claim in the report should come from a page you opened.
Method
-
Plan before searching. Decompose the question into three to six specific sub-questions and state them for the user in one short line each. Vague sub-questions produce vague queries produce vague reports.
-
Search narrowly, once per sub-question. One well-formed query each. If a query returns nothing, broaden it -- drop a
site:filter, a quoted phrase, or a year -- rather than rephrasing it sideways. Repeating a near-identical query is the most common way a research turn burns its budget and still arrives with nothing. -
Open the sources. For each sub-question,
read_urlthe two or three most promising results. Prefer primary sources: the company's own filing, the standards document, the paper, the maintainer's changelog. Prefer them over an article summarizing them, and over a forum thread describing them. -
Note what you could not establish. A page that would not load, a number with only one source behind it, a claim every result traces back to the same press release. These go in the report. A research report whose whole value is that it can be trusted has to be explicit about where it thins out.
-
Synthesize, then cite. Write from the page text you gathered. Link every non-obvious claim inline as a markdown link to the specific page it came from -- not to the site's home page, and not to the search results. Where sources disagree, say so and give both.
Shape of the deliverable
- TL;DR -- the answer in three to five sentences, up front.
- Sections -- one per sub-question, headed, with inline links.
- Tables where the content is comparative.
- What we could not establish -- the gaps and single-sourced claims from step 4.
- Sources -- the pages you actually opened, not everything the search returned.
Honesty
Never present a figure, date, or quotation you did not read on a page. If the research did not reach a conclusion, the report says so; it does not round an absence up into a finding. If live search is unavailable this turn, say that plainly and do not write the report from memory as though you had searched.
Zeplik output presentation
Present the final deliverable as a single polished artifact: clear headings, tables where the content is tabular, fenced code where it is code. Lead with the deliverable itself; keep process commentary to a single short line. If the skill produced multiple files or sections, end with a compact list of them with one-line purposes.
How to use the Deep Research Reports skill
Sign in to Zeplik
Create a free Zeplik account or sign in. New accounts start with free credits, so you can try the Deep Research Reports skill right away.
Describe your research task
Ask in plain language, or type /deep-research to invoke the skill directly. Zeplik recognizes the Deep Research Reports skill and applies its method.
Review and refine the result
Zeplik returns a structured document you can edit, download, and reuse. Ask follow-ups to refine it.
Source and credit
- Author
- davila7 community
- License
- MIT
Adapted from the open-source davila7/claude-code-templates project and tuned to run natively on Zeplik. View source on GitHub.
Frequently asked questions
- What is the Deep Research Reports skill?
- Deep Research Reports is a ready-to-run research skill on Zeplik. Plan, search, open the real pages and synthesize a long multi-step cited report -- market analysis, competitive landscaping, literature review, due diligence, comparisons. Ask in plain language and Zeplik applies the skill's method for you inside the conversation, on whichever AI model you prefer. It returns a structured document you can keep and reuse: Document artifact -- structured written deliverable with headed sections and a TL;DR (see artifact-templates/document.md).
- How do I use Deep Research Reports on Zeplik?
- Sign in to Zeplik and ask in plain language, or type /deep-research in any chat to invoke it directly. The skill applies its method and returns a result you can refine in the same conversation.
- Which AI model does the Deep Research Reports skill use?
- Any model you choose. Zeplik works across every model in one chat, so the Deep Research Reports skill runs on your preferred model for the task.
- Where does the Deep Research Reports skill come from?
- The Deep Research Reports skill is adapted from the open-source davila7/claude-code-templates project (MIT) and tuned to run natively on Zeplik. The original source is linked on this page.
- How much does the Deep Research Reports skill cost?
- Using the skill is free to start. You only spend Zeplik credits when the assistant runs, and new accounts begin with free credits.
Related research skills
- Academic PublishingAcademic publishing workflows — citation/reference management, LaTeX research posters, and venue templates with submission requirements; for the writing itself see research-writing.
- Bio Research ToolkitUse when running comp-bio workflows -- single-cell RNA-seq QC, scvi-tools deep models, nf-core pipelines, instrument-to-Allotrope conversion. Not for literature research (use deep-research).
- Cheminformatics ToolkitsCheminformatics and molecular modeling — RDKit/Datamol, DeepChem molecular ML, COBRApy metabolic modeling, Pymatgen materials, matchms/pyOpenMS mass spec; genomics: genomics-toolkits.
- Clinical WritingClinical and medical document generation — clinical decision-support docs, clinical/case reports (CARE guidelines), and focused treatment plans in LaTeX/PDF; manuscripts: research-writing.
- Competitive BriefResearch named competitors and write the brief: positioning, messaging, product, content, market presence, gaps and threats. For a live-call battlecard see competitive-intelligence.
- Fact CheckerCheck whether a claim is true and say so with evidence -- verify against live primary sources and grade true, mostly true, misleading, unsupported or false, with citations.
More on Zeplik
Try Deep Research Reports on Zeplik
Every model, one chat. Bring the Deep Research Reports skill into your next conversation and let the assistant do the work.