Recursive Llm Ts

B 82 completed
Library
monorepo / typescript · small
161
Files
29,475
LOC
2
Frameworks
11
Languages

Pipeline State

completed
Run ID
#348759
Phase
done
Progress
1%
Started
Finished
2026-04-13 01:31:02
LLM tokens
0

Pipeline Metadata

Stage
Cataloged
Decision
proceed
Novelty
68.13
Framework unique
Isolation
Last stage change
2026-05-10 03:34:51
Deduplication group #49482
Member of a group with 2 similar repo(s) — this repo is canonical view group →
Top concepts (2)
Project DescriptionTesting
Repobility — same analyzer, your code, free for public repos · /scan/

AI Prompt

Create a TypeScript library that implements Recursive Language Model (RLM) functionality to handle unbounded context processing for LLMs. I need features for structured output generation using Zod schemas, streaming results via async iterables, and batch operations with concurrency control. The tool should support multiple providers like OpenAI, Anthropic, and Azure, and ideally include context overflow recovery strategies like mapreduce or chunked processing. Please ensure the API is highly typed and includes a builder pattern for configuration.
typescript llm ai context-window structured-output async-iterable openai zod monorepo
Generated by gemma4:latest

Catalog Information

A TypeScript library that enables recursive language model usage for unlimited context processing and structured output generation.

Description

This library provides a TypeScript bridge to a recursive language model framework, allowing developers to process text that exceeds typical token limits by recursively summarizing or chunking content. It exposes a simple API for initiating recursive calls, managing context windows, and parsing structured outputs such as JSON or custom schemas. Validation of inputs and outputs is handled through a schema library, ensuring type safety and consistency. The tool is designed for AI developers who need to build applications that can handle large documents, long conversations, or complex data transformations. It integrates seamlessly with existing TypeScript projects and supports testing via a popular unit testing framework.

الوصف

توفر هذه المكتبة واجهة TypeScript لدمج نماذج اللغة المتكررة، ما يتيح للمطورين معالجة نصوص تتجاوز حدود عدد الرموز المعتادة عن طريق تلخيص أو تقسيم المحتوى بشكل متكرر. تُقدّم واجهة برمجية بسيطة لبدء الاستدعاءات المتكررة، وإدارة نوافذ السياق، وتحليل المخرجات المنظمة مثل JSON أو مخططات مخصصة. يتم التحقق من صحة المدخلات والمخرجات عبر مكتبة تعريفات، مما يضمن أمان النوعية والاتساق. صممت هذه الأداة للمطورين في مجال الذكاء الاصطناعي الذين يحتاجون إلى بناء تطبيقات تتعامل مع مستندات طويلة، محادثات مستمرة، أو تحويلات بيانات معقدة. تتكامل بسلاسة مع مشاريع TypeScript الحالية وتدعم اختبارات الوحدة عبر إطار عمل شائع.

Novelty

7/10

Tags

recursive-processing context-management structured-output llm-integration data-validation

Technologies

vitest zod

Claude Models

claude-opus-4.5 claude-opus-4.6

Quality Score

B
82.4/100
Structure
90
Code Quality
79
Documentation
94
Testing
85
Practices
63
Security
82
Dependencies
60

Strengths

  • Well-documented README with substantial content
  • CI/CD pipeline configured (github_actions)
  • Good test coverage (67% test-to-source ratio)
  • Code linting configured (ruff (possible))
  • Good security practices \u2014 no major issues detected
  • Containerized deployment (Docker)
  • Properly licensed project

Weaknesses

  • 1 bare except/catch blocks swallowing errors
  • Potential hardcoded secrets in 1 files
  • 995 duplicate lines detected \u2014 consider DRY refactoring
  • 3 'god files' with >500 LOC need decomposition

Recommendations

  • Replace bare except/catch blocks with specific exception types
  • Move hardcoded secrets to environment variables or a secrets manager

Security & Health

8.6h
Tech Debt (A)
A
OWASP (100%)
PASS
Quality Gate
A
Risk (1)
Powered by Repobility — scan your code at https://repobility.com
MIT
License
6.1%
Duplication
Full Security Report AI Fix Prompts SARIF SBOM

Languages

typescript
24.1%
go
22.2%
markdown
16.6%
html
12.6%
python
8.4%
json
7.0%
css
3.2%
javascript
2.1%
yaml
1.9%
shell
1.9%
toml
0.2%

Frameworks

pytest Vitest

Concepts (2)

Source-of-truth: Repobility · https://repobility.com
CategoryNameDescriptionConfidence
Generated by Repobility's multi-pass static-analysis pipeline (https://repobility.com)
auto_descriptionProject DescriptionTypeScript/JavaScript package for Recursive Language Models (RLM) - process unbounded context lengths with LLMs.80%
auto_categoryTestingtesting70%

Quality Timeline

1 quality score recorded.

View File Metrics

Embed Badge

Add to your README:

![Quality](https://repos.aljefra.com/badge/72848.svg)
Quality BadgeSecurity Badge
Export Quality CSVDownload SBOMExport Findings CSV