# Clouditera/SecGPT

SecGPT网络安全大模型

Repository: https://github.com/Clouditera/SecGPT
Canonical: https://ross.abutalabs.com/products/secgpt
Language: Python
License: Apache-2.0
License Family: permissive
Last push: 2025-06-25T10:03:37+00:00

## Health v2 (maintenance only)
Score: 39/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 28, release rhythm 35, longevity 72
- inputs: {"age_days": 1017, "days_push": 434, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 3101, forks 368 (observed 2026-08-28T04:07:43.578257+00:00)

## What it is
SecGPT is an open-source large language model purpose-built for cybersecurity, fine-tuned from Qwen2.5 and DeepSeek-R1 bases on a large curated security corpus. The repository provides model weights, training details, and deployment instructions via vLLM for security analysis tasks.

## Use cases
- analyze vulnerabilities and generate remediation advice with an LLM
- trace attack paths from logs and network traffic
- detect anomalies and potential security threats
- support red team and blue team decision making
- parse attack scripts and identify malicious commands
- answer cybersecurity knowledge questions
- self-host a security-focused chat model with vLLM

## When to choose
- you need an LLM specialized in security operations, vulnerability analysis, or threat intelligence
- you want a self-hosted security assistant running on your own GPUs
- you need a lightweight security model (SecGPT-Mini) that runs on CPU

## When to avoid
- you need a general-purpose assistant unrelated to cybersecurity
- you lack GPU resources for the full-size model and cannot use the mini variant
- you require guaranteed accuracy for critical security decisions without human review

## Facets
- artifact type: learning-resource
- maturity: active
- function: llm-inference, llm-training, nlp, security, machine-learning
- domain: security, large-language-models, artificial-intelligence
- platform: python, self-hosted
- tags: cybersecurity, open-source-llm, vulnerability-analysis, threat-detection, fine-tuned-model, qwen, deepseek, vllm, security-qa, red-team, natural-language-processing, gpu, linux, docker

## Member repositories
- Clouditera/SecGPT (main) score 39

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:07:43.578257+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T07:26:59.162387+00:00, confidence not recorded.
  - readme: https://github.com/Clouditera/SecGPT (fetched 2026-08-28T04:07:43.578257+00:00, sha edf6b07571ec)
- Data as of 2026-08-30T08:39:29.467469+00:00.