Research Lookup for Manuscripts
Gathers and organizes scholarly sources and evidence for a scientific manuscript or research brief.
- Repository
- K-Dense-AI/scientific-agent-skills
- Stars
- 55,780
- Licence
- MIT
- Files
- 5 files
What it does
When you explicitly ask for literature or background evidence for a scientific manuscript, this skill tells the agent to run a bundled Python script that searches the web through Parallel's search service, then retrieves the text of promising sources. It organizes what it finds into a research packet with a reference ledger and a claim-to-source style evidence summary, rather than a loose list of links. It counts a reference as verified only when its text was successfully retrieved, and it still expects you to check that each source truly supports your claim and has not been corrected or retracted.
Great for
- Researchers who need a starting set of recent papers, reviews and competing findings for a manuscript introduction or discussion
- Graduate students building a background evidence base they will then read and verify themselves
- Writers who want a structured packet of sources with DOIs and PMIDs to review
Skip it if
- Systematic reviews or PRISMA-style work, which need protocols, database-specific searches and screening; the skill itself points to a different skill for that (reason: it cannot guarantee completeness)
- Anyone expecting verified citations without review: retrieval success does not confirm bibliographic identity, claim support, or retraction status (reason: risk of citing wrong or unsupported papers)
- Private or unpublished data, since query text and context are sent to external services (reason: confidentiality)
Try asking
- “Use the research lookup skill to gather 60 recent peer-reviewed sources on intervention X and adverse events for my manuscript”
- “Find competing or null findings on this claim and build an evidence matrix I can review”
- “Compile a manuscript research packet for my study using the context file I provide”
Install in 3 steps
- 1
Download the skill
The ZIP contains the 5 original files, the MIT licence and an attribution note.
research-lookup.zip - 2
Add it to your agent
Claude Code (all projects)
mkdir -p ~/.claude/skills && unzip research-lookup.zip -d ~/.claude/skills/Claude Code / Agent SDK (this project only)
mkdir -p .claude/skills && unzip research-lookup.zip -d .claude/skills/Claude apps (web/desktop): Settings → Capabilities → Skills → Upload skill, and choose the ZIP.
- 3
Ask for the task
No special command needed — the agent reads the skill's description and loads it when your request matches. Try one of the prompts above.
Get better results
- Provide a context file with your research question, population, exposure, comparator and outcomes so searches stay focused; do not let it invent study details
- Check the cost and scope of the academic run first, since it makes many paid search and extract calls, and use --extract-limit to reduce spend
- Treat every source as a lead: read each excerpt, confirm DOI/PMID identity, and check corrections and retractions at the publisher before citing
What's inside
- README.md
- references/api-contracts.md
- scripts/manuscript_packet.py
- scripts/research_lookup.py
- SKILL.md
Redistributed unmodified from K-Dense-AI/scientific-agent-skills under the MIT licence. The assessment was drafted with AI from the skill's files on 2026-10-29. Skills can instruct your agent to run code — read SKILL.md and any scripts before installing.