datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
agentskillos-benchmark
AgentSkillOS Benchmark
Official benchmark release for Organizing, Orchestrating, and Benchmarking Agent Skills at Ecosystem Scale.
GitHub project: https://github.com/ynulihao/AgentSkillOS
This release is a file-based benchmark package. It contains the original task definitions and the bundled task assets that are needed by tasks with local input files.
Contents
data/tasks.jsonl: normalized task index for load_dataset(...)
tasks/: the original 30 benchmark task… See the full description on the dataset page: https://huggingface.co/datasets/NPULH/agentskillos-benchmark.agent-skill-vulnerabilities
Agent Skill Vulnerability Scenarios (defanged, teaching)
from datasets import load_dataset
ds = load_dataset("fevziegeyurtsevenler/agent-skill-vulnerabilities")
Deliberately-vulnerable, defanged agent-extension artifacts — for training detectors & hands-on learning.
Schema
column
meaning
id
scenario
artifact_type
SKILL.md / mcp.json
content, walkthrough
artifact + defense
Related AltaySec resources
🕵️ uncloak scanner:… See the full description on the dataset page: https://huggingface.co/datasets/fevziegeyurtsevenler/agent-skill-vulnerabilities.atr-skill-benchmark
ATR Skill-Security Benchmark
A labeled corpus of SKILL.md files for evaluating detection of malicious agent
skills — prompt injection, tool poisoning, credential theft, malware droppers
and supply-chain attacks hidden inside natural-language agent instructions.
Published as part of Agent Threat Rules (ATR),
an open, vendor-neutral detection standard for AI agents (like Sigma, but for
agent attacks).
Why this exists
SKILL.md files are natural-language instructions… See the full description on the dataset page: https://huggingface.co/datasets/Agent-Threat-Rule/atr-skill-benchmark.agent-skill-malware
Agent Skill Malware: Malicious vs Benign Agent Skills
Binary classification dataset of OpenClaw agent skill files (SKILL.md).
127 malicious + 223 benign = 350 samples, deduplicated by content hash.
Source
Malicious: Real skills from malicious campaigns targeting ClawHub (Feb 2026), extracted from the openclaw/skills GitHub archive. They use social engineering in markdown instructions to trick agents/users into running malware -- primarily AMOS (Atomic macOS Stealer)… See the full description on the dataset page: https://huggingface.co/datasets/yoonholee/agent-skill-malware.han-cross-agent-skill-transfer-benchmark-dataset-v1
Humanoid Cross-Agent Skill Transfer Benchmark Dataset
This dataset benchmarks how effectively
skills learned by one humanoid agent
can be transferred to another agent
within a decentralized cognitive network.
Objective
To measure cross-agent generalization,
adaptation speed, and transfer efficiency.
Data Fields
source_agent_skill_profile
target_agent_initial_profile
transferred_skill_vector
adaptation_steps
performance_improvement_percentage… See the full description on the dataset page: https://huggingface.co/datasets/achiepatricia/han-cross-agent-skill-transfer-benchmark-dataset-v1.agent-skill-malware
Agent Skill Malware: Malicious vs Benign Agent Skills
Binary classification dataset of OpenClaw agent skill files (SKILL.md).
127 malicious + 223 benign = 350 samples, deduplicated by content hash.
Source
Malicious: Real skills from malicious campaigns targeting ClawHub (Feb 2026), extracted from the openclaw/skills GitHub archive. They use social engineering in markdown instructions to trick agents/users into running malware -- primarily AMOS (Atomic macOS Stealer)… See the full description on the dataset page: https://huggingface.co/datasets/TongxiQu/agent-skill-malware.agent-skills