Aumai Metaagent
B+ 86 completed
Ai Ml
cli / markdown · tiny
22
Files
3,061
LOC
1
Frameworks
4
Languages
Pipeline State
completedRun ID
#304018Phase
doneProgress
1%Started
Finished
2026-04-13 01:31:02LLM tokens
0Pipeline Metadata
Stage
CatalogedDecision
proceedNovelty
46.39Framework unique
—Isolation
—Last stage change
2026-05-10 03:35:02Deduplication group #51033
Member of a group with 16 similar repo(s) — canonical #5452 view group →
Top concepts (4)
Project DescriptiontestingTestingTesting
Repobility · severity-and-effort ranking · https://repobility.com
🧪 Code Distillation
Browse all specs →Sample distilled functions (click for full spec)
EvolutionConfigDefines a configuration structure for controlling an evolutionary meta-learning process. It accepts various hyperparameters such as population size, total generations, and mutation rates as inputs. The structure itself serves as a data container, outputting a validated set of parameters for the evol
PerformanceMetricCreates a structured data object to record a specific performance evaluation for an agent blueprint. It accepts a blueprint identifier, and optional floating-point scores for fitness and mean episode reward, a success rate, an integer count for evaluation steps, an integer for the generation number,
parameter_countCalculates an estimate of the total number of trainable parameters within the model's hidden layers. It accepts no inputs and returns an integer representing the calculated parameter count. The function iterates through the stored hidden layer sizes, accumulating the connections between successive l
AI Prompt
Create a command-line agent, similar to AumAI Metaagent, that is designed to learn how to build better agents. The project should be written primarily in Python and use pytest for testing. Include documentation sections for getting started, an API reference, and examples. Since it's a CLI tool, please structure it to be easily installable via pip.
python cli agent ai pytest markdown tooling
Generated by gemma4:latest
Catalog Information
This project is an agent that learns to build better agents.
Description
Aumai-metaagent is a system designed to create and improve artificial intelligence agents. It uses machine learning techniques to learn from existing agents and generate new, more effective ones. The goal of this project is to develop a meta-agent that can autonomously design and optimize AI systems.
الوصف
هذا المشروع عبارة عن एजنت تعلم كيفية بناء أجنة أفضل. يعتمد على تقنيات التعلم الآلي لتعلم من الأجنة الموجودة وإنشاء أجنة جديدة أكثر فعالية. الهدف من هذا المشروع هو تطوير أجنة متعددة يمكنها تصميم وتحسين أنظمة الذكاء الإصطناعي بشكل تلقائي.
Novelty
9/10Tags
artificial-intelligence machine-learning meta-agent agent-design optimization
Technologies
click numpy pydantic
Claude Models
claude-opus-4.6
Quality Score
B+
85.8/100
Structure
88
Code Quality
90
Documentation
85
Testing
85
Practices
68
Security
100
Dependencies
90
Strengths
- CI/CD pipeline configured (github_actions)
- Good test coverage (60% 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
Security & Health
4.1h
Tech Debt (C)
High
DORA Rating
A
OWASP (100%)
Repobility · code-quality intelligence · https://repobility.com
PASS
Quality Gate
A
Risk (3)
Apache-2.0
License
0.0%
Duplication
Languages
Frameworks
pytest
Symbols
variable27
constant12
method11
class7
function4
Concepts (4)
| Category | Name | Description | Confidence | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Repobility · code-quality intelligence · https://repobility.com | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| auto_description | Project Description | > Agent that learns to build better agents | 80% | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| arch_layer | testing | Detected testing layer | 70% | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| auto_category | Testing | testing | 70% | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| business_logic | Testing | Detected from 3 related files | 50% | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Powered by Repobility — scan your code at https://repobility.com
Embed Badge
Add to your README:
BinComp Dependency Hardening
All packages →2 of this repo's dependencies have been scanned for binary hardening. Grade reflects RELRO / stack canary / FORTIFY / PIE coverage.