ClawKit Logo
ClawKitReliability Toolkit
Back to Registry
Official Verified

feishu-openclaw-paper-manager

Design or implement a paper-management workflow built on a Feishu bot plus OpenClaw. Use when the user wants to ingest papers from Feishu messages, save PDFs or source links into a Feishu cloud-docs paper folder, maintain a searchable multi-dimensional table of paper metadata and reusable tags, or evolve the tag taxonomy after every 50 newly added papers.

skill-install — Terminal

Install via CLI (Recommended)

clawhub install openclaw/skills/skills/ch1hyaanon/feishu-paper-manager
Or

Feishu OpenClaw Paper Manager

Use this skill when the task is to design, review, or implement a paper-management agent that runs through a Feishu bot and OpenClaw.

Primary objective

Build a workflow with three durable outputs:

  1. A cloud-docs paper folder that stores the paper asset or an index document for the source link.
  2. A Feishu multi-dimensional table that stores normalized paper metadata.
  3. A taxonomy iteration loop that improves labels whenever total paper count crosses 50, 100, 150, and so on.

Workflow

Follow this sequence:

  1. Map the ingestion sources from Feishu messages.
  2. Normalize each paper into one canonical record.
  3. Save or register the paper in the cloud-docs paper folder.
  4. Generate summary and reusable multi-label tags.
  5. Write the record into the multi-dimensional table.
  6. Check whether the total record count has reached a multiple of 50.
  7. If yes, run taxonomy refinement and backfill historical rows.

Ingestion rules

Treat these as valid paper inputs:

  • PDF attachment in a Feishu message
  • arXiv, OpenReview, ACL Anthology, publisher, or project links in a message
  • Mixed messages that contain both a PDF and a source link

For every incoming item:

  1. Extract the raw message URL, sender, timestamp, and all detected paper candidates.
  2. Resolve whether the message refers to a new paper or an existing one.
  3. Create exactly one canonical paper record per paper.

Deduplication priority:

  1. DOI
  2. arXiv ID or OpenReview forum ID
  3. normalized title
  4. source URL fingerprint

If a duplicate exists, update missing fields instead of creating a new row.

Storage model

Always separate binary storage from metadata storage:

  • The cloud-docs paper folder is for the PDF file or a small link-index doc when the PDF is unavailable.
  • The multi-dimensional table is the system of record for metadata, classification, and retrieval.

When a PDF is available:

  • upload the PDF into the paper folder
  • use a deterministic filename from year + first_author + short_title

When only a link is available:

  • create a lightweight cloud doc in the paper folder containing the title, source link, access date, and capture notes
  • still create the metadata row in the table

Required table fields

Create or expect these fields in the Feishu multi-dimensional table:

Metadata

Stars3875
Views0
Updated2026-04-07
View Author Profile
AI Skill Finder

Not sure this is the right skill?

Describe what you want to build — we'll match you to the best skill from 16,000+ options.

Find the right skill
Add to Configuration

Paste this into your clawhub.json to enable this plugin.

{
  "plugins": {
    "official-ch1hyaanon-feishu-paper-manager": {
      "enabled": true,
      "auto_update": true
    }
  }
}
Safety NoteClawKit audits metadata but not runtime behavior. Use with caution.