by jordan-gibbs · Claude Skill · ★ 248
The Most Powerful Deep Research Harness Hyperresearch turns Claude Code into a deep research agent. and currently leads the DeepResearch-Bench RACE leaderboard (benchmarked internally). A tier-adaptive 16-step pipeline produces adversarially-audited reports with full source provenance. Every fetched source lands in a persistent, searchable vault that compounds across sessions. Forward-looking projection from a stratified pilot against the DeepResearch-Bench leaderboard snapshot (https://huggingface.co/spaces/muset-ai/DeepResea
| Stars | 248 |
| Forks | 16 |
| Language | Python |
| Category | Claude Skill |
| License | MIT |
| Quality Score | 67.4012709088015/100 |
| Last Updated | 2026-05-14 |
| Created | 2026-04-09 |
| Platforms | browser, claude-code, python |
| Est. Tokens | ~72k |
These tools work well together with hyperresearch for enhanced workflows:
Explore other popular claude skill tools:
hyperresearch is Agent-driven research knowledge base. Agents collect, search, and synthesize web research into a persistent, searchable wiki.. It is categorized as a Claude Skill with 248 GitHub stars.
hyperresearch is primarily written in Python. It covers topics such as agents, agentskills, claude-code.
You can find installation instructions and usage details in the hyperresearch GitHub repository at github.com/jordan-gibbs/hyperresearch. The project has 248 stars and 16 forks, indicating an active community.
hyperresearch is released under the MIT license, making it free to use and modify according to the license terms.