SkillRLaiming-lab/SkillRL
Builds a recursively evolving hierarchical skill bank for AI agents and uses reinforcement learning to let agents accumu…
No ratings yet
Rate this skill
Stars1k
Forks80
Watchers1k
Open Issues1
InstallOne-line install
UnconfirmedWhether this repository follows the Agent Skills spec (that is, whether it contains a SKILL.md) has not been verified yet. Until that is settled, clone it and use what you need:
git clone https://github.com/aiming-lab/SkillRL.git
OverviewOverview
Key Features
Builds a recursively evolving hierarchical skill bank for AI agents and uses reinforcement learning to let agents accumulate and refine skills during tasks. Key capabilities include: 1. Hierarchical SKILLBANK: organizes skills into a layered bank with retrieval and reuse to reduce repeated learning cost; 2. Recursive Skill Evolution: recursively generates and iterates new skills from task feedback so the bank keeps expanding during training; 3. Context Efficiency: selects high-value skills and information within limited context windows to cut redundant input; 4. SFT Data Generation: provides code for supervised fine-tuning data generation used in skill learning and model training experiments.
Use Cases
agent skill bank construction, RL training data construction, automatic skill evolution experiments, context compression and efficiency optimization, model fine-tuning and evaluation.
Target Roles
AI / Algorithm Engineer, Backend Developer, Data Analyst / Data Engineer.
RatingUser ratings
Nobody has rated this skill yet. Yours would be the first on the site — your vote becomes the initial average.
RepositoryRepository
| Skill ID | 1038 |
|---|---|
| Data source | GitHub |
| Repository | aiming-lab/SkillRL |
| Category | AI & Machine Learning / Reinforcement Learning & Agent Training |
| Primary language | Python |
| License | MIT License |
| Default branch | main |
| GitHub repository ID | 1152962622 |
| Node ID | R_kgDORLjQPg |
| Rank | #762 (position within the source ranking) |
| Site-wide popularity rank | #862 (computed live from Stars, across all 5833 entries) |
GrowthStar / Fork growth
One snapshot per day, tracking how Stars and Forks changed
- Now1,004
- Previous point · 09-29 990
- vs previous point +14
- 2 data points 12 days
- Now80
- Previous point · 09-29 78
- vs previous point +2
- 2 data points 12 days
TimelineTimeline
Repository created2026-02-08 17:40:219 months ago
Data last updated2026-10-26 12:03:3021 days ago
Last commit2026-09-25 14:18:571 months ago
CommitLatest commit
Committed by richard-peng-xia. Update Category: New Features & Capability Expansion. Commit details: Add news about SkillRL acceptance at NeurIPS 2026. This update introduces new capabilities and expands supported use cases.