Human-Agent-Society/reef
Infrastructure for continually self‑improving agents
View on GitHub →PythonApache-2.0agent-infrastructureai-agentscontinual-learninginferencellm
Stars
7.8k
Δ 7d
+42
Δ 30d
+896
Age
1 mo
Last push
1h ago
90-day star history
Where this sits in the catalog
GroupFrameworks & SDKs
RankingsNot in any top 20 today — see the rankings
Most-wanted open issues
Ranked by 👍 reactions. 77 open issues total, checked 2026-10-25.
| Issue | 👍 | Comments | Opened |
|---|---|---|---|
[Community] Reef Roadmap roadmap, status: keep-open | 5 | 0 | 1 mo |
[Feature] Add a Self-Distillation Fine-Tuning (SDFT) recipe for continual learning from demonstrations enhancement, area: examples, area: training, status: ready | 1 | 0 | 1 mo |
[RFC] Evaluation-calibrated adaptive KL for online policy updates enhancement, area: training, RFC, status: ready | 1 | 4 | 1 mo |
[Task] Split the "Harness Adapter" page area: docs, type: task, status: ready | 0 | 1 | 2 wk |
[Bug] Reporting an inference record without policy tensors stops automatic training bug, area: core, status: needs-triage | 0 | 0 | 6 d |
Alternatives in Agent SDKs & frameworks
Agent skills that give coding agents taste: art direction, reference images and a browser check instead of gen…
⭐104.0k
+3,8387d
JavaScript
8 mo
Open-source personal AI assistant & Agent Harness. Plans tasks, runs tools and skills, self-evolves with memor…
⭐47.6k
+1517d
Python
4 yrs