ZETSU is a personal offline RAG system that ingests offensive security writeups and notes, then allows natural language querying for commands, attack chains, and concepts.

Stars

162

7-day growth

No data

Forks

9

Open issues

0

License

No data

Last updated

2026-07-13

AI repository intelligence
FR-AI / ANALYSIS

Why it is worth attention

It combines hybrid retrieval (BM25+vector), cross-encoder reranking, and a structured extraction format (FARR) tailored for offensive security workflows, all while keeping data entirely offline and private.

Who it is for

  • Penetration testers and red teamers
  • Bug bounty hunters
  • Security researchers and students
  • Offensive security professionals managing personal knowledge bases

Use cases

  • Quickly retrieving specific commands from past writeups during an engagement
  • Cross-referencing multiple sources (HTB writeups, blogs, tool wikis) for a single technique
  • Studying attack chains by querying structured FARR entries
  • Automating knowledge retrieval via CLI for scripting and reporting

Strengths

  • Hybrid retrieval (BM25 + vector search) with RRF fusion ensures both exact and semantic matching
  • Cross-encoder reranker improves relevance of chunks sent to the LLM
  • FARR extraction converts writeups into structured attack steps (Finding, Action, Reasoning, Result)
  • Supports multiple LLM backends (Anthropic, OpenAI, Ollama) and two response styles (Operator vs Concept)

Considerations

  • ChromaDB loads full vector index into RAM; large corpora (>10k chunks) may require remote ChromaDB server (planned)
  • Quality of answers depends heavily on the user's own notes and sources ingested
  • Requires local setup and configuration; not a cloud-based turnkey solution

README quick start

Installation

Description

A personal RAG system for offensive security knowledge

Related repositories

Similar projects matched by category, topics, and programming language.

lopopolo
Featured
lopopolo GitHub avatar

harness-engineering

Harness Engineering is a methodology for improving coding agent outputs by carefully crafting the environment around them—providing curated context, tools, and executable constraints that encode an organization’s nonfunctional requirements and cumulative lessons.

AI & Machine LearningAI Agents
2,390
slvDev
Featured
slvDev GitHub avatar

esp32-ai

A 28.9 million parameter language model runs on an $8 ESP32-S3 microcontroller entirely on-device, generating simple stories at about 9.5 tokens per second.

AI & Machine LearningLarge Language Models
1,960
littledivy
Featured
littledivy GitHub avatar

mimic

mimic captures traffic from any iOS or web app and automatically generates a Python client library that lets you call the app's API like a regular library.

AI & Machine Learning
1,482