---
license: "mit"
pretty_name: "OpenSkillRisk"
language:
  - en
tags:
  - agent-safety
  - coding-agents
  - skills
  - benchmark
gated: true

extra_gated_heading: "Dataset Access Request"
extra_gated_description: "Access requests are reviewed by the dataset maintainers. Please allow several business days for review."
extra_gated_button_content: "Submit access request"

extra_gated_fields:
  "Full Name": text
  "Affiliation": text
  "Institutional or Work Email": text
  "Intended Use":
    type: select
    options:
      - "Academic research"
      - "Safety evaluation"
      - "Education or teaching"
      - "Other"
  "I will not attempt to identify original skill authors or related organizations": checkbox
  "I will not redistribute the raw skill files": checkbox
---

# OpenSkillRisk Dataset

## Dataset Description

OpenSkillRisk is a benchmark for evaluating whether CLI agents can safely handle risky third-party skills in open skill ecosystems.

This dataset contains the skill files and task specifications used by OpenSkillRisk. The code for building sandboxes and running evaluations is released separately at:

- **Repository:** [Miaow-Lab/OpenSkillRisk](https://github.com/Miaow-Lab/OpenSkillRisk)
- **Paper:** [OpenSkillRisk: Benchmarking Agent Safety When Using Real-World Risky Third-Party Skills](https://arxiv.org/abs/2607.20121)

## Dataset Structure

This dataset contains two top-level folders:

- `skills/` contains the third-party skill files used in the benchmark.
- `tasks/` contains the task specification files that define benchmark instances.

## Dataset Size

| Split | Number of skills and tasks |
| --- | ---: |
| Obviously malicious | 139 |
| Contextually risky | 124 |
| Total | 263 |


| Risk category | Total number of skills and tasks |
| --- | ---: |
| Control-Plane Hijacking | 13 |
| Authority Expansion | 38 |
| Data Harvesting | 27 |
| Execution Bootstrapping | 30 |
| Persistence Implantation | 24 |
| Outbound Exfiltration | 96 | 
| External State Manipulation | 35 | 


## Usage

Please refer to the GitHub repository for instructions on downloading this dataset and running OpenSkillRisk.


## Important Notice

The third-party skills in this repository contains high-risk skills collected for controlled safety research and benchmark evaluation. These skills should not be redistributed, copied into other agent environments, or used outside the intended benchmark context.

The benchmark is designed for sandboxed evaluation, but users should still run it only in controlled, security-conscious environments and avoid exposing sensitive credentials or personal data.

## License

MIT License. See `LICENSE`.

## Citation

If you use OpenSkillRisk in your research, please cite:

```bibtex
@article{liu2026openskillrisk,
  title={OpenSkillRisk: Benchmarking Agent Safety When Using Real-World Risky Third-Party Skills},
  author={Liu, Qiyuan and Hui, Tingfeng and Zhan, Kun and Zhang, Kaike and Miao, Ning},
  journal={arXiv preprint arXiv:2607.20121},
  year={2026}
}
```

## Takedown and Removal Requests

If any skill, file, or derived content in this dataset belongs to you and you would like it removed, please provide a brief description of your ownership or rights claim and contact us. We will review valid requests and update the dataset as appropriate.