by HarryHsing · Agent Tool · ★ 70
Last updated: · Indexed by AgentSkillsHub · Auto-synced every 8h
OmniAgent (ICML 2026): the first native omni-modal agent for active video perception — a 7B agent that beats Qwen2.5-VL-72B with 73% fewer frames on LVBench.
| Stars | 70 |
| Forks | 6 |
| Language | Python |
| Category | Agent Tool |
| License | Apache-2.0 |
| Quality Score | 54.2963281311495/100 |
| Open Issues | 2 |
| Last Updated | 2026-09-02 |
| Created | 2026-06-11 |
| Platforms | python |
| Est. Tokens | ~13k |
These tools work well together with OmniAgent for enhanced workflows:
Looking for a OmniAgent alternative? If you're comparing OmniAgent with other agent tool tools, these 6 projects are the closest alternatives on Agent Skills Hub — ranked by topic overlap, star count, and community traction.
Awesome LLM Papers and repos on very comprehensive topics.
NEWTON: Agentic Planning for Physically Grounded Video Generation
[Up-to-date] A curated list of resources on graph-empowered agents and agent-facilitated graph learning (Graph
基于多模态视觉感知与 LLM Agent 的 macOS 微信自动化框架 | Visual RPA for WeChat
[NeurIPS 2024] ReEvo: Large Language Models as Hyper-Heuristics with Reflective Evolution
✨✨Latest Advances on Neuro-Symbolic Learning in the era of Large Language Models
Explore other popular agent tool tools:
OmniAgent is OmniAgent (ICML 2026): the first native omni-modal agent for active video perception — a 7B agent that beats Qwen2.5-VL-72B with 73% fewer frames on LVBench.. It is categorized as a Agent Tool with 70 GitHub stars.
OmniAgent is primarily written in Python. It covers topics such as agent, audio-visual, icml2026.
You can find installation instructions and usage details in the OmniAgent GitHub repository at github.com/HarryHsing/OmniAgent. The project has 70 stars and 6 forks, indicating an active community.
OmniAgent is released under the Apache-2.0 license, making it free to use and modify according to the license terms.
The top alternatives to OmniAgent on Agent Skills Hub include Awesome-LLM-Papers-Comprehensive-Topics, NEWTON, Awesome-Graphs-Meet-Agents. Each offers a different approach to the same problem space — compare them side-by-side by stars, quality score, and community activity.
Grades come from a rule-based scan built on the SlowMist agent-security taxonomy, covering 11 red-flag categories including credential harvesting, data exfiltration, and curl | sh installers. It is a first-layer scan, not a manual audit — we say so rather than overstate it.
The scale of the problem is documented independently: Liu et al. (2026), in a study of 31,132 agent skills, report that 26.1% contain security vulnerabilities. Our own full-catalog census is published as a citable open dataset.
Sources & who's responsible: