Paper Radar

A research habit, with less sorting

A shorter reading list.
A reason for every pick.

Describe your research interest or share paper links. Let the skill retrieve public titles and abstracts, then get a ranked page, reasons and files you can keep.

MIT licensed · No separate API key for conversation mode · Normal Claude plan usage applies

A conversation reportReal papers

RESEARCH INTEREST

LLM agent benchmarks and tool use

2 abstracts2 shortlisted

Must-read0.93

AgentBench: Evaluating LLMs as Agents

Direct match: evaluates language models as agents in interactive environments.

Maybe0.74

Toolformer: Language Models Can Teach Themselves to Use Tools

Related method: teaches models to use tools, rather than introducing an agent benchmark.

Saved test with two real papers. Claude used labelled web summaries after the metadata API was blocked. Scores are self-reported, uncalibrated and may vary; this page does not search or rank live.

Start where you are

Look first. Then make it yours.

02

Start with an interest or links

Upload and enable skill 0.2.0, then give a research interest or arXiv/PubMed/DOI links. Public retrieval needs network access or existing Claude web tools. The conversation ranks and exports the results.

Install and get a prompt →

From download to first result

Your first shortlist
starts with one interest.

  1. Download the complete skill ZIP.Use paper-radar-claude-v0.2.0.zip, not the repository Source code ZIP.
  2. Upload it in Claude → Customize → Skills.Enable paper-radar and code execution/file creation. An organization may restrict these controls.
  3. Start a chat with the prompt on the right.Try the fictional demo, describe an interest, or paste supported paper links. CSV is optional.
Download Claude skill ZIP ↓Read the full installation guideTry two real links: first-run guide →
Open Claude ↗

This page only prepares a prompt; it sends nothing. In Claude, public retrieval sends the query or identifiers to metadata sources or permitted web tools. Missing abstracts and failures are disclosed.

Keep the result, not just the chat

One reading list. Four useful files.

Links above open the saved real-paper test, using labelled web summaries. HTML/CSV/JSON retain all scored papers; RSS includes must-read and maybe. The original fictional report remains available in the repository.

Before your first run

Know what you are getting.

Do I need an API key?

No separate model API key for conversational retrieval, ranking or exports. Normal Claude plan usage applies. Retrieval makes public metadata requests and needs permitted network or existing web tools. Scheduled model API judging is separately billed.

Does installation start a daily radar?

No. You can request a single search in the current conversation. Daily schedules, long-term storage and publishing still need separate setup.

Is a score of 0.93 a 93% chance of relevance?

No. Claude's scores are self-reported and uncalibrated. Label selected and unselected papers yourself, fit thresholds on one batch and evaluate on a separate later batch.

What has actually been checked?

Claude web Opus 5.5 loaded the bundle, checked its file hashes and exported reports. The 0.2.0 link test retrieved two real papers through existing Claude web tools after the metadata API was blocked; summaries were labelled. Local tests: 250; Python 3.11–3.13 CI passed. Functional checks do not establish accuracy or every-client compatibility.

What should I share when something goes wrong?

Your environment, package version, the step that failed and a redacted error. Do not post keys, private abstracts or unpublished research. A first-run report is more useful than a star.