OpenAgentSkill
GitHub200+

Skill rankings

Best document processing skills for AI agents

Find skills for parsing PDFs, extracting tables, running OCR, converting documents, and preparing file content for agent workflows.

Choose a ranking

Shown: 5 · Candidates: 62

Compare top 4

Saved directory data is shown because current data is unavailable. Dates and metrics may be out of date.

  1. 01

    MarkItDown

    Convert PDFs, Office documents, and web files into clean markdown for agents.

    @microsoftDocument Processing80,000
    Review before use
    View details
    Ranking signals

    Signals use a 0–100 scale. They explain the shortlist; the ordering follows this list’s method, not any one score.

    Popularity
    98/100
    Quality
    100/100
    Freshness
    0/100
    Agent evidence
    0/100
    Evidence confidence
    0/100
    Install readiness
    90/100
    Task fit
    30/100
  2. 02

    Canvas Design

    Create original visual art, posters, PNG assets, and PDF documents through a clear design philosophy.

    @anthropicsDesign163,076
    Review before use
    View details
    Ranking signals

    Signals use a 0–100 scale. They explain the shortlist; the ordering follows this list’s method, not any one score.

    Popularity
    100/100
    Quality
    100/100
    Freshness
    0/100
    Agent evidence
    0/100
    Evidence confidence
    0/100
    Install readiness
    100/100
    Task fit
    23/100
  3. 03

    Firecrawl

    Turn websites into clean markdown or structured data for retrieval and agents.

    @mendableaiWeb Scraping34,000
    Review before use
    View details
    Ranking signals

    Signals use a 0–100 scale. They explain the shortlist; the ordering follows this list’s method, not any one score.

    Popularity
    91/100
    Quality
    100/100
    Freshness
    0/100
    Agent evidence
    0/100
    Evidence confidence
    0/100
    Install readiness
    90/100
    Task fit
    20/100
  4. 04

    Crawl4AI

    Open-source LLM-friendly web crawler and scraper for agent workflows.

    @unclecodeWeb Scraping66,000
    Review before use
    View details
    Ranking signals

    Signals use a 0–100 scale. They explain the shortlist; the ordering follows this list’s method, not any one score.

    Popularity
    96/100
    Quality
    100/100
    Freshness
    0/100
    Agent evidence
    0/100
    Evidence confidence
    0/100
    Install readiness
    90/100
    Task fit
    17/100
  5. 05

    LlamaIndex

    Data framework for building RAG and knowledge workflows around agent tasks.

    @run-llamaRAG42,000
    Review before use
    View details
    Ranking signals

    Signals use a 0–100 scale. They explain the shortlist; the ordering follows this list’s method, not any one score.

    Popularity
    93/100
    Quality
    100/100
    Freshness
    0/100
    Agent evidence
    0/100
    Evidence confidence
    0/100
    Install readiness
    90/100
    Task fit
    16/100

A shortlist from up to 480 directory candidates, not the entire registry. Stars belong to repositories. Signals are not safety guarantees or runtime verification.

How this list works

Matches task and source metadata, then considers quality, popularity and freshness. Source descriptions are not execution evidence.

Signals use a 0–100 scale. They explain the shortlist; the ordering follows this list’s method, not any one score.