Repository: CRUD_RAG
Author: IAAR-Shanghai·Source status: Clear source
CRUD-RAG: A Comprehensive Chinese Benchmark for Retrieval-Augmented Generation of Large Language Models
Score basis：Clear source · High execution risk · Universal · Evidence completeness 67%
Pre-install score
90 · Manual review
One primary score for ranking and pre-install decisions; Evidence completeness 67%.
Risk decision
Review required
metadata-only
Install readiness
script-backed · copy-only command
SkillTrust only shows install guidance and copy actions; it never executes installs.
Before you install
Review source, permissions, and execution risk first, then alternatives. Scores prioritize review; they do not replace manual judgment.
Install guidance
This area handles the execution step: review explicit or inferred tool fit, then decide whether to copy commands.
Explicitly supported
Universal
Shown with the common install pattern for this tool.
git clone https://github.com/IAAR-Shanghai/CRUD_RAG.gitReview source and permissions first. If no explicit command is available, start from the repository guidance.
Current risk hints: metadata-only
Audit main stage
This stage turns 15 audit dimensions into one pre-install score; the constellation only explains the gaps instead of becoming a second score.
The purpose is visible, but source, permissions, or install evidence needs checking.
Current advice
Review before install
Review source, permissions, and execution details before copying commands.
Execution profile
High
Check command, file-write, and network behavior first.
Evidence completeness
67%
This shows how much public evidence supports the score; it is not a safety certification.
Start with the overall shape and the weakest three items, then use the list below for full labels, scores, and actions.
Start here
Read the weakest three first, then move into the full 15-item list.
Full 15-dimension list
The default view starts from the lowest scores; switch to risk-first when you want severity before score.
Repository
IAAR-Shanghai/CRUD_RAG
Author
IAAR-Shanghai
Community signal
370 stars · 30 forks
Favorites
0
Last updated
2025-05-20
Primary source
IAAR-Shanghai/CRUD_RAG
Source status
Clear source
Install method
script-backed
Related: Same task category...
Why related: Same task category, Keyword overlap, Similar install method
Repository: graphrag
Author: microsoft · Source status: Clear source
A modular graph-based Retrieval-Augmented Generation (RAG) system
Score basis：Clear source · High execution risk · Universal · Evidence completeness 67%
Related: Same task category...
Why related: Same task category, Keyword overlap, Similar install method
Repository: FastGPT
Author: labring · Source status: Clear source
FastGPT is a knowledge-based platform built on the LLMs, offers a comprehensive suite of out-of-the-box capabilities such as data processing, RAG retrieval, and visual AI workflow orchestration, letting you easily devel…
Score basis：Clear source · High execution risk · Universal · Evidence completeness 67%
Related: Same task category...
Why related: Same task category, Keyword overlap, Similar install method
Repository: RAG_Techniques
Author: NirDiamant · Source status: Clear source
This repository showcases various advanced techniques for Retrieval-Augmented Generation (RAG) systems.
Score basis：Clear source · High execution risk · Universal · Evidence completeness 67%
Related: Same task category...
Why related: Same task category, Keyword overlap, Similar install method
Repository: Awesome-AI-Memory
Author: IAAR-Shanghai · Source status: Clear source
Awesome AI Memory | LLM Memory | A curated knowledge base on AI memory for LLMs and agents, covering long-term memory, reasoning, retrieval, and memory-native system design.
Score basis：Clear source · High execution risk · Universal · Evidence completeness 67%
Repository: Awesome-Attention-Heads
Author: IAAR-Shanghai · Source status: Clear source
An awesome repository & A comprehensive survey on interpretability of LLM attention heads.
Score basis：Clear source · High execution risk · Universal · Evidence completeness 65%