Skillgraph

C+ 72 completed
Cli Tool
library / python · tiny
24
Files
1,848
LOC
1
Frameworks
3
Languages

Pipeline State

completed
Run ID
#343600
Phase
done
Progress
1%
Started
Finished
2026-04-13 01:31:02
LLM tokens
0

Pipeline Metadata

Stage
Cataloged
Decision
proceed
Novelty
45.49
Framework unique
Isolation
Last stage change
2026-05-10 03:26:32
Deduplication group #48300
Member of a group with 9 similar repo(s) — canonical #74290 view group →
Top concepts (2)
Project DescriptionTesting
All rows scored by the Repobility analyzer (https://repobility.com)

AI Prompt

Create a security scanner, similar to SkillGraph, for analyzing AI agent skill files. I need it to be a Python library that can parse Markdown skills into structured data and perform pattern-based risk detection to identify malicious behaviors like data exfiltration or credential theft. The tool should offer a command-line interface for scanning and parsing, and ideally, include a Streamlit-based web UI for visualizing the potential risks using an interactive knowledge graph. Please ensure it handles different risk levels (Critical, High, Medium, Low).
python security scanner cli streamlit markdown ai-agent risk-detection library
Generated by gemma4:latest

Catalog Information

A security scanner that analyzes AI agent skill files for hidden malicious behavior and visualizes potential risks.

Description

SkillGraph is a command‑line and web‑based tool that scans AI agent skill files written in Markdown for hidden malicious commands. It parses the skill structure, applies pattern‑based rules to detect risks such as data exfiltration or destructive actions, and builds a knowledge graph to illustrate relationships between components. The CLI produces quick reports in text or JSON, while the Streamlit web UI offers interactive visualizations of the risk graph. Targeted at AI developers and security analysts, it helps ensure that downloaded skills are safe before deployment, addressing the problem of undisclosed threats in third‑party code.

الوصف

يُعد SkillGraph أداة فحص أمان للمهارات الخاصة بالذكاء الاصطناعي، حيث يتيح للمستخدم تحليل ملفات المهارات المكتوبة بالـ Markdown قبل استخدامها. تقوم الأداة بتحليل بنية الملف واستخراج العناصر الأساسية مثل الأوامر والوظائف. ثم تُطبق مجموعة من القواعد النمطية لاكتشاف مخاطر محتملة مثل استخراج البيانات أو تنفيذ أوامر تدمير النظام. تُبنى شبكة رسومية توضح العلاقات بين المكونات المختلفة للمهارة، ما يسهل على المستخدم رؤية نقاط الضعف المحتملة. تُعرض النتائج عبر واجهة سطر أوامر لتوليد تقارير سريعة، كما يمكن تشغيل تطبيق ويب لتصور الرسوم البيانية بشكل تفاعلي. تُستهدف هذه الأداة مطوري الذكاء الاصطناعي ومحللي الأمان الذين يحتاجون إلى ضمان سلامة المهارات قبل نشرها.

Novelty

7/10

Tags

ai-agent-security risk-detection markdown-parsing graph-analysis cli-tool web-visualization

Technologies

anthropic langchain openai pytorch streamlit

Claude Models

claude-opus-4.6

Quality Score

C+
71.8/100
Structure
64
Code Quality
87
Documentation
65
Testing
50
Practices
67
Security
100
Dependencies
60

Strengths

  • Consistent naming conventions (snake_case)
  • Good security practices \u2014 no major issues detected

Weaknesses

  • No LICENSE file \u2014 legal ambiguity for contributors
  • No CI/CD configuration \u2014 manual testing and deployment

Recommendations

  • Set up CI/CD (GitHub Actions recommended) to automate testing and deployment
  • Add a linter configuration to enforce code style consistency
  • Add a LICENSE file (MIT recommended for open source)

Security & Health

5.1h
Tech Debt (D)
A
OWASP (100%)
PASS
Quality Gate
A
Risk (7)
About: code-quality intelligence by Repobility · https://repobility.com
Unknown
License
0.0%
Duplication
Full Security Report AI Fix Prompts SARIF SBOM

Languages

python
69.2%
markdown
30.2%
text
0.6%

Frameworks

pytest

Concepts (2)

Open data · scored by Repobility · https://repobility.com
CategoryNameDescriptionConfidence
Repobility (the analyzer behind this table) · https://repobility.com
auto_descriptionProject Description> Map the Hidden Risks - AI Agent Skills Security Scanner80%
auto_categoryTestingtesting70%

Quality Timeline

1 quality score recorded.

View File Metrics

Embed Badge

Add to your README:

![Quality](https://repos.aljefra.com/badge/67650.svg)
Quality BadgeSecurity Badge
Export Quality CSVDownload SBOMExport Findings CSV