AI Skill MartBrowse skillsPromptsProgramsLearnWhat is a Skill?
 
← Back to list

Research Lookup (Manuscript Evidence Packet)

Builds a manuscript-ready research packet — 60 verified, deduplicated references plus evidence matrix and BibTeX — using Parallel Search, Extract, and Research.

EducationIntermediate★ 33,030⑂ 3,248AI score 9/10Last updated: Aug 9, 2026

What it does

  • Takes manuscript context (research question, study type, population, exposure, comparator, outcomes, date range) via a JSON file to constrain retrieval.
  • Runs six bounded academic search passes: recent primary studies, systematic reviews/meta-analyses, seminal work, methods and protocols, contradictory/null evidence, and an unrestricted companion pass to close gaps.
  • Verifies candidates with Parallel Extract for authors, year, venue, DOI/PMID, design, sample size, quantitative findings, limitations, and retraction status.
  • Writes an auditable packet directory: packet.json/md, references.json/.bib, evidence-matrix.json, claim-source-map.json, synthesis.json, section-briefs.json, coverage.json, search-ledger.json.
  • Enforces citation hygiene: dedupe by DOI/PMID/title, exclude retracted work, flag preprints, never fabricate metadata, never pad shortfalls.

Who it's for

  • Grad students and researchers drafting Introduction, Methods rationale, and Discussion sections.
  • Anyone who needs both supporting and conflicting evidence for a scientific claim.
  • Teams that want verified reference metadata and BibTeX rather than a pile of links.
  • Not for PRISMA systematic reviews — hand those to the literature-review skill.

Example uses

  1. Create manuscript-context.json, then run the script with --academic --target-references 60 --packet-dir sources/manuscript-research --json for a full packet.
  2. On an explicit "do exhaustive research" request, use --force-backend research --processor pro -o sources/deep-research.md.
  3. When the reference target is missed, open coverage.json and refine terminology, date range, or domains instead of lowering quality.
  4. Cover several questions at once with --batch "query one" "query two", each getting its own packet subdirectory.

· · · Install guide · · ·

Try it now, no install

Paste this into Claude to use the skill without installing anything.

Read the instructions in this file and follow them to help me:
https://raw.githubusercontent.com/K-Dense-AI/scientific-agent-skills/HEAD/skills/research-lookup/SKILL.md

What I want: (describe your task here)

If Claude can't open the link, open it yourself and paste the contents instead.

↓ If it works for you, download the ZIP below and install it. Then it runs on its own — no pasting each time.

Install in the Claude app (no terminal)
  1. Download the ZIP with the button below.
  2. In Claude, open Settings → Capabilities and turn on 'Code execution and file creation'. (one time)
  3. Go to Customize → Skills → + → 'Upload a skill' and upload the ZIP.
↓ Download ZIP
Install in Claude Code

Let Claude do it — paste this into Claude Code

Install the skill I found on AI Skill Mart.
Copy the skills/research-lookup folder from the GitHub repo K-Dense-AI/scientific-agent-skills into my ~/.claude/skills/research-lookup/.
When it's done, tell me in one line what this skill can do.

Install with a command instead

git clone https://github.com/K-Dense-AI/scientific-agent-skills.git && mkdir -p ~/.claude/skills && cp -r scientific-agent-skills/skills/research-lookup ~/.claude/skills/

⚠ This is a third-party skill. Check the source repository before installing.

  1. Clone the repo: git clone https://github.com/K-Dense-AI/scientific-agent-skills.git
  2. Install the skill: mkdir -p ~/.claude/skills && cp -r scientific-agent-skills/skills/research-lookup ~/.claude/skills/
  3. Install the pinned CLI: uv tool install "parallel-web-tools[cli]==0.7.1" (install uv first if you don't have it).
  4. Authenticate: parallel-cli login, or parallel-cli login --device on headless machines.
  5. For the Chat backend or headless runs, export PARALLEL_API_KEY in your shell profile; add OPENROUTER_API_KEY only if you want the optional Perplexity path. Never pass keys as command arguments.
  6. Verify with parallel-cli --version and parallel-cli auth.
  7. Restart Claude Code and ask something like "gather literature and build a manuscript research packet for this research question."
View source on GitHub ↗License: MIT