agentleFS
lucaswychanlucaswychan/quant-paper-agent/.claude/skills/paper-search/SKILL.md

paper-search

Skill0 starsChanged 3 months agoMITView on GitHub

Search arXiv quantitative-finance categories, resolve exact paper versions, and download reproducible PDF source bundles. Use for q-fin paper discovery, date/category/keyword searches, arXiv IDs or URLs, reading queues, and batch downloads that must preserve source and discovery provenance for later extraction or replication.

18 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use extract-paper-images, paper-analyze and 10 more.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • Clay-HHKSkill

    extract-paper-images

    从论文中提取图片,优先从arXiv源码包获取真正的论文图

    Also in 5 projects that use this

  • Clay-HHKSkill

    paper-analyze

    深度分析单篇论文,生成详细笔记和评估,图文并茂

    Also in 4 projects that use this

  • Google CloudSkill

    citation-verification

    Verify that citations in a document resolve to real records and that claimed titles match. Use before trusting a document's reference list.

    Also in 3 projects that use this

  • MicrosoftSkill

    web-design-reviewer

    This skill enables visual inspection of websites running locally or remotely to identify and fix design issues. Triggers on requests like "review website design", "check the UI", "fix the layout", "find design problems". Detects issues with responsive design, accessibility, visual consistency, and layout breakage, then performs fixes at the source code level.

    Also in 3 projects that use this

  • xfstudioSkill

    code-review-excellence

    Master effective code review practices to provide constructive feedback, catch bugs early, and foster knowledge sharing while maintaining team morale. Use when reviewing pull requests, establishing review standards, or mentoring developers.

    Also in 3 projects that use this

  • xfstudioSkill

    uv-package-manager

    Master the uv package manager for fast Python dependency management, virtual environments, and modern Python project workflows. Use when setting up Python projects, managing dependencies, or optimizing Python development workflows with uv.

    Also in 3 projects that use this

  • +2 more from Clay-HHK

The file

We split this skill into the rules it gives an agent

5 rules. 0 are common, 2 are distinctive, and 3 are very specific to this project.

  • 0

    Common

    Most of the 11 edited versions we read keep these.

  • 2

    Distinctive

    This author's own choices. Worth a look.

  • 3

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Create an exact, auditable source bundle.

    line 8
  2. Very specific:

    Read references/search_manifest.schema.json when consuming saved search results and references/source_manifest.schema.json when checking a …

    line 34
  3. Distinctive:

    Use HTTPS and preserve the exact version.

    line 51
  4. Very specific:

    Treat truncated: true as incomplete discovery.

    line 54
  5. Very specific:

    Do not parse PDFs here. paper-extract owns content inspection.

    line 59

“Common” is measured: kept by at least 6 in 10 of the 11 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Paper search

Create an exact, auditable source bundle. Do not extract or summarize the PDF in this stage.

Resolve paths

Resolve the absolute workspace root and this skill's directory before running anything. Invoke bundled scripts by absolute path; never assume the workspace contains a top-level scripts/search_arxiv.py.

QPA_SEARCH_SKILL_DIR="/absolute/path/to/paper-search"
QPA_WORKSPACE_ROOT="/absolute/path/to/workspace"
python3 "$QPA_SEARCH_SKILL_DIR/scripts/search_arxiv.py" \
  --query "cross-sectional momentum" --category q-fin.PM \
  --since 2024-10-01 --max 10 \
  --output "$QPA_WORKSPACE_ROOT/searches/momentum-2024/results.json"

Review broad results before downloading. The sidecar results.manifest.json records the exact query, response checksums, total result count, pagination, and truncation state. Writers validate both manifest types against their bundled current JSON Schemas before saving.

Download only selected records:

python3 "$QPA_SEARCH_SKILL_DIR/scripts/download_paper.py" 2403.12345v2 \
  --root "$QPA_WORKSPACE_ROOT/papers"

Artifact contracts

Read references/search_manifest.schema.json when consuming saved search results and references/source_manifest.schema.json when checking a source bundle.

searches/<search-id>/
├── results.json
└── results.manifest.json

papers/<versioned-id>/
├── paper.pdf
├── meta.json
└── source_manifest.json

An unversioned request must resolve to a versioned arXiv identity. New bundles record PDF and metadata checksums. A verified cached bundle is not rewritten.

Rules

  • Use HTTPS and preserve the exact version.
  • Validate IDs and dates before requesting them.
  • Save discovery manifests for searches that inform a reading queue or download decision.
  • Treat truncated: true as incomplete discovery; narrow the query or page deliberately.
  • Download atomically and reject non-PDF responses.
  • Reuse a cached PDF only when its checksum agrees with its manifest.
  • Respect arXiv throttling, retries, and the configured identifying user agent.
  • Surface empty searches and network errors; never invent papers or widen scope silently.
  • Do not parse PDFs here. paper-extract owns content inspection.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.