Strands Sglang

C+ 79 completed
Ai Ml
unknown / python · small
55
Files
6,676
LOC
1
Frameworks
4
Languages

Pipeline State

completed
Run ID
#303032
Phase
done
Progress
1%
Started
Finished
2026-04-13 01:31:02
LLM tokens
0

Pipeline Metadata

Stage
Skipped
Decision
skip_scaffold_dup
Novelty
49.77
Framework unique
Isolation
Last stage change
2026-04-16 18:15:42
Deduplication group #48287
Member of a group with 1 similar repo(s) — canonical #6293 view group →
Top concepts (6)
RepositoryProject DescriptiontestingTestingAuthenticationTesting
Repobility · open methodology · https://repobility.com/research/

AI Prompt

Create a Python library that integrates the SGLang model provider with the Strands Agents SDK. The main goal is to enable agent-based reinforcement learning training by providing token-in/token-out rollouts. Specifically, the implementation must support token IDs along with their corresponding logprobs and masks to prevent retokenization drift. I'm looking for a solution that helps make the Strands Agents SDK training-ready by exposing these end-to-end, token-level rollouts.
python sglang reinforcement-learning agent-sdk rl-training token-out pytorch
Generated by gemma4:latest

Catalog Information

The strands-sglang project provides a SGLang model to the Strands Agents SDK, enabling token-in/token-out support for agent-based reinforcement learning training.

Description

Strands-sglang is a model provider designed for use with the Strands Agents SDK. It supports token-in and token-out functionality, which is essential for agent-based reinforcement learning training. This project leverages the capabilities of the SGLang model to facilitate efficient and effective training processes.

الوصف

هذا المشروع يوفّر نموذج SGLang للمكتبة Strands Agents SDK، ويوفر دعمًا لوصف الإدخال والخروج بالتوقيعات، مما يساعد على تدريب العمليات التعلمية المستندة إلى الوكلاء.

Novelty

7/10

Tags

agent-based-reinforcement-learning rl-training model-provider sglang-model strands-agents-sdk

Technologies

huggingface openai pytorch

Claude Models

claude-opus-4.6 claude-opus-4.5

Quality Score

C+
78.6/100
Structure
89
Code Quality
75
Documentation
64
Testing
85
Practices
72
Security
90
Dependencies
90

Strengths

  • CI/CD pipeline configured (github_actions)
  • Good test coverage (125% test-to-source ratio)
  • Code linting configured (ruff (possible))
  • Consistent naming conventions (snake_case)
  • Good security practices \u2014 no major issues detected
  • Properly licensed project

Weaknesses

  • Potential hardcoded secrets in 1 files

Recommendations

  • Move hardcoded secrets to environment variables or a secrets manager

Security & Health

4.6h
Tech Debt (B)
High
DORA Rating
A
OWASP (100%)
Open data scored by Repobility · https://repobility.com
PASS
Quality Gate
A
Risk (2)
Apache-2.0
License
11.5%
Duplication
Full Security Report AI Fix Prompts SARIF SBOM

Languages

python
89.9%
markdown
7.2%
yaml
1.9%
toml
1.0%

Frameworks

pytest

Symbols

method44
variable22
class21
constant16
function13
property13

Concepts (6)

Scored by Repobility's multi-pass pipeline · https://repobility.com
CategoryNameDescriptionConfidence
All rows scored by the Repobility analyzer (https://repobility.com)
design_patternRepositoryFound repository-named files80%
auto_descriptionProject Description![CI](https://github.com/horizon-rl/strands-sglang/actions/workflows/test.yml) ![PyPI](https://pypi.org/project/strands-sglang/) ![License](LICENSE)80%
arch_layertestingDetected testing layer70%
auto_categoryTestingtesting70%
business_logicAuthenticationDetected from 4 related files50%
business_logicTestingDetected from 25 related files50%

Quality Timeline

1 quality score recorded.

View File Metrics
Powered by Repobility — scan your code at https://repobility.com

Embed Badge

Add to your README:

![Quality](https://repos.aljefra.com/badge/26856.svg)
Quality BadgeSecurity Badge
Export Quality CSVDownload SBOMExport Findings CSV

BinComp Dependency Hardening

All packages →
4 of this repo's dependencies have been scanned for binary hardening. Grade reflects RELRO / stack canary / FORTIFY / PIE coverage.
Ntransformers5.5.3 · 0 gadgets · risk 10146.5Faiohttp3.13.5 · 354 gadgets · risk 8061.4Nasyncio4.0.0 · 0 gadgets · risk 0.0Npydantic2.12.5 · 0 gadgets · risk 0.0