Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →brycewang-stanford/paper-recommendation-guide
Systematic paper recommendation and discovery using multiple methods
Passed 20 of 24 test cases with the skill, and 17 of 24 without it.
and 2 more in the full report
5 fixed15 passed either way2 still failing2 broke
Tokens, turns and time describe resource use, not a dollar price.
One of 5 test cases the skill fixed. Same model, same prompt, with and without the skill.
I have accumulated over 200 unread PDFs in my reference manager over the past year. What specific operational rules should I implement to prevent an ever-growing reading queue, including reading time-boxing and queue expiration rules?
| Checked | Without skill | With skill |
|---|---|---|
| Recommends time-boxing specific schedule blocks for dedicated paper reading. | Passed | Passed |
| Recommends archiving queue papers that remain unread after 8 weeks. | Missed | Passed |
The instructions your model reads when the skill loads.
Finding the right papers to read is a research skill in itself. Beyond keyword searches, modern researchers have access to a rich ecosystem of recommendation tools that use citation networks, semantic similarity, co-authorship patterns, and collaborative filtering to surface relevant papers you might otherwise miss.
This skill provides a systematic approach to paper discovery that goes beyond passive reading. It covers algorithmic recommendation services, citation-based discovery techniques, social and community-driven methods, and strategies for building and maintaining a well-curated reading pipeline. The goal is to minimize the chance that you miss an important paper while avoiding information overload.
Whether you are entering a new field and need foundational papers, tracking the frontier of a mature research area, or looking for interdisciplinary connections, this guide provides concrete methods for each scenario.
OpenAlex provides concept-based and citation-based discovery for 250M+ works across all disciplines:
# Find works related to a specific paper via its concepts and citations
curl "https://api.openalex.org/works?filter=cites:W2741809807&sort=cited_by_count:desc&per_page=10"
Use OpenAlex's concept graph to find related work by browsing papers tagged with the same research concepts, or trace citation networks to find derivative and foundational papers.
Connected Papers (connectedpapers.com) builds a visual graph of papers related to a seed paper. It uses co-citation and bibliographic coupling analysis rather than direct citation links, which means it can surface related work even when two papers do not cite each other directly. Use this when:
Google Scholar's "Related articles" feature and the personalized recommendation emails (if you maintain a Google Scholar profile) use a combination of citation analysis and content similarity. To maximize their usefulness:
test cases passed, vs 17 / 24 without
More skills from the same source and nearby in the registry, with their headline benchmark lift.