Stable Baselines3 skill: what it does and how to install it
Trains and evaluates single-agent reinforcement-learning agents with Stable Baselines3 across custom Gymnasium environments, vectorised rollouts and common algorithms.
Summary generated from the skill's documentation.
Install
$ npx skills add K-Dense-AI/scientific-agent-skills --skill stable-baselines3Run it in a terminal. If your agent is already running, start a new session so it picks the skill up.
About this skill
What it does. Guides setup and use of Stable Baselines3 for training and evaluating single-agent reinforcement-learning agents. It covers algorithm selection, custom Gymnasium environments, vectorised environments, callbacks, checkpoints, normalisation, evaluation and video recording, with templates and reference guidance.
When to use it. Applies to reproducible reinforcement-learning experiments involving continuous control, discrete actions, custom environments, or SB3-Contrib recurrent and masked policies.
History
Repo stars
47.7kAbout +18.9k since 9 Jul 2026
Before 1 Oct 2026 the curve is estimated from public event data.
Stars are counted for the whole repository, which holds 74 skills.
Show as a table
| Date | Repo stars |
|---|---|
| 5 Oct 2026 | 47,652 |
| 4 Oct 2026 | 47,541 |
| 3 Oct 2026 | 47,444 |
| 2 Oct 2026 | 47,351 |
| 1 Oct 2026 | 47,271 |
| 24 Sept 2026 (estimated) | 46,976 |
| 17 Sept 2026 (estimated) | 46,257 |
| 10 Sept 2026 (estimated) | 41,836 |
| 3 Sept 2026 (estimated) | 30,545 |
| 27 Aug 2026 (estimated) | 29,347 |
| 20 Aug 2026 (estimated) | 29,160 |
| 13 Aug 2026 (estimated) | 29,027 |
| 6 Aug 2026 (estimated) | 28,947 |
| 30 Jul 2026 (estimated) | 28,894 |
| 23 Jul 2026 (estimated) | 28,894 |
| 16 Jul 2026 (estimated) | 28,841 |
| 9 Jul 2026 (estimated) | 28,708 |
Installs
1.7k
Tracking since . A chart appears once there are 7 days of data.
Installs via skills.sh
Similar skills
- Dispatching Parallel AgentsSplits two or more independent tasks across parallel agents when they share no state.
- Karpathy GuidelinesBehavioural guidelines that reduce common LLM coding mistakes: avoid overcomplication, make surgical changes and state assumptions.
- Caveman ExploreExplores a repository read-only to locate code across files and returns path and line citations.