SkillAgentSearch skills...

awesome-paper2agent

A community collecting paper MCP packages — each one wraps a paper's code and workflow as an MCP server that an agent can install and call.

Install / Use

claude mcp add omicverse -- npx -y github:omicverse/awesome-paper2agent

If the server publishes to npm under a different name, use that package instead — check the repo README.

About this skill
🔌

MCP Server

Model Context Protocol server

Quality Score

78/100

Category

Security

Supported Platforms

Claude Code
Claude Desktop

Tags

Our assessment of awesome-paper2agent

awesome-paper2agent scores 78/100 on our quality scale, 1033rd of 1,130 Security skills we index.

Its MCP Server is 4.6 KB long, well organised into 8 sections with 4 code examples: a solid amount of guidance for an agent.

It has 3 GitHub stars, so there is little community track record yet; judge it on its content.

Substance
26/30
Structure
20/20
Description
15/15
Adoption
3/20
Freshness
15/15

Maintenance, license and trust

  • The repository was last updated 28 days ago, so awesome-paper2agent is actively maintained.
  • It is released under the Apache-2.0 license, a permissive license that allows use, modification and commercial use with attribution.
  • Its trust signals score 92/100, with 1 caution from licensing, adoption, age or documentation. These come from repository metadata, not a code audit — read the skill file before letting an agent act on it.

Safety scan

No issues found

Our scan of the whole file found no instruction hijacking, hidden characters, credential access, data exfiltration or destructive commands.

Automated pattern scan on 2026-10-23. It catches known dangerous patterns, not every risk — read a skill before letting an agent act on it.

awesome-paper2agent compared with similar skills

All 4 of these similar skills score higher than awesome-paper2agent; compare them before choosing.

SkillScoreStarsUpdatedFormat
awesome-paper2agent (this skill)by omicverse78328d agoMCP Server
Agent-Reachby Panniantong100101.0k2d agoCLAUDE.md
headroomby headroomlabs-ai10077.5ktodayCLAUDE.md
CowAgentby zhayujie10047.5ktodayCLAUDE.md
Scraplingby D4Vinci10089.2ktodayMCP Server

Frequently asked questions

How do I install awesome-paper2agent?
Run claude mcp add omicverse -- npx -y github:omicverse/awesome-paper2agent. The install tabs above show the steps for each supported agent.
Which AI agents does awesome-paper2agent work with?
It is written for Claude Code and Claude Desktop, as a MCP Server file. Other agents that read the same format can often use it too.
Is awesome-paper2agent safe to use?
Our scan of the whole file found no instruction hijacking, hidden characters, credential access, data exfiltration or destructive commands. It is Apache-2.0-licensed and scores 92/100 on trust signals. Skills are instructions an agent will follow, so read the file before installing it and do not approve commands you do not understand.
Is awesome-paper2agent still maintained?
The repository was last updated 28 days ago, so awesome-paper2agent is actively maintained.

awesome-paper2agent

License Metadata schema v1 Contributions: PR

🇨🇳 中文版: README.zh.md

A paper's value extends beyond being read, yet its methods are often dispersed across code, notebooks and supplementary material. Making use of them requires understanding the implementation, aligning dependencies and running the workflow end to end. That is a barrier to use, and it also means the same packaging gets repeated by different people.

We built the Paper2Agent community to collect finished paper MCP packages in one place; once installed in OmicOS, a paper's methods become callable in the conversation like any built-in tool.


Quick start

Paste this into any agent, attach the paper, and it will build a package that passes validation and is ready to open as a pull request:

Follow AGENTS.md in this repository to build a package from the attached paper PDF,
using <upstream repository URL>. Focus on <methods or figures to expose as tools>.

Build instructions for agents live in AGENTS.md — one file, one source of truth.

How a package reaches users

<picture> <source media="(prefers-color-scheme: dark)" srcset="docs/assets/flow-en-dark.svg"> <img src="docs/assets/flow-en-light.svg" width="677" alt="Source → Pull request → Static checks → Maintainer review → ZIP + index → MCP client"> </picture>

Every published package records the paper it comes from, the upstream repository, the exact commit, its license, and the checksum of the artifact. reviews.json pins maintainer approvals to package content, so any change to code or metadata requires a new review. A package stays out of the release index until it is approved, and the default branch refuses to release while an unapproved package sits under packages/.

What is in this repository

| Path | Purpose | |---|---| | packages/<package_id>/ | Submitted packages; published once approved | | examples/sequence-stats/ | A synthetic example that exercises the full catalog pipeline | | schema/package-v1.schema.json | Versioned metadata contract for submissions | | tools/catalog.py | Validates and deterministically packages sources | | tools/from_paper2mcp.py | Converts an OmicOS build-pipeline delivery into a package | | docs/contract.md | Consumer contract, artifact layout and limitations | | AGENTS.md | Build instructions for agents |

<details> <summary>Package layout</summary>
packages/<package_id>/
  metadata.json          strict schema; no fields outside it
  USAGE.md               purpose, inputs, outputs, limits, install steps
  VALIDATION.md          how each tool was verified, and what was not
  LICENSE                license of this submission
  NOTICE                 optional upstream notices
  src/requirements.txt   exactly pinned dependencies (package==version)
  src/<name>_mcp.py      the single MCP entry point
</details>

Local checks

python3 -m venv .venv && . .venv/bin/activate
pip install -r requirements-dev.txt
python -m unittest discover -s tests -v
python tools/catalog.py validate
python tools/catalog.py demo          # local example catalog in dist/demo/
python tools/catalog.py build --revision $(git rev-parse HEAD)

validate --require-approved additionally fails while a package under packages/ has no matching entry in reviews.json. The required release-gate check runs it on pull requests and pushes. A release build requires a committed, clean source revision.

Demo runtime check

requirements-dev.txt covers validation only. The example server needs its own environment:

python3.11 -m venv .runtime-venv
.runtime-venv/bin/pip install -r examples/sequence-stats/src/requirements.txt
PYTHONDONTWRITEBYTECODE=1 .runtime-venv/bin/python tools/smoke_example.py

Licensing

The repository's own contents — schema, validator, workflows and documentation — are Apache-2.0; see LICENSE. A submission is not relicensed by it: packages/<package_id>/LICENSE stays authoritative for that package, and the declared license must permit redistribution of everything it contains. See CONTRIBUTING.md.

Contributing

Contributions are submitted as pull requests and reviewed before merge. Merging code and approving content are separate: an approval in reviews.json is required before a package is published.

Related Skills

View on GitHub
GitHub Stars3
CategorySecurity
Updated28d ago
Forks1

Languages

Python

Trust signals

92/100

From repository metadata: license, adoption, age and documentation. Not a code audit — see the Safety scan above for what the skill file itself contains.

1 low