# Acmesec/PromptJailbreakManual

Prompt越狱手册

Repository: https://github.com/Acmesec/PromptJailbreakManual
Canonical: https://ross.abutalabs.com/products/promptjailbreakmanual
License: GPL-3.0
License Family: copyleft
Last push: 2024-12-17T13:11:24+00:00

## Health v2 (maintenance only)
Score: 12/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 0, release rhythm 8, longevity 46
- inputs: {"age_days": 646, "days_push": 624, "days_rel": 645, "gap_med": null, "n_releases_24m": 1}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 3608, forks 374 (observed 2026-08-28T04:08:11.598291+00:00)

## What it is
A Chinese-language manual (handbook) covering prompt engineering, prompt injection, prompt leaking, and LLM jailbreak techniques, with security-oriented case studies. It is educational documentation rather than software.

## Use cases
- learn prompt engineering techniques
- understand prompt injection attacks
- study LLM jailbreak methods
- build a security-focused AI assistant prompt
- learn prompt frameworks like LangGPT and COAST
- red-team LLM safety testing

## When to choose
- you want a structured reference on prompt design and jailbreak techniques
- you do security research or red-teaming of LLM applications
- you want real-world prompt injection and leaking case studies

## When to avoid
- you need executable tooling or a library
- you want an English-language resource
- you need formal academic coverage of LLM safety

## Facets
- artifact type: learning-resource
- maturity: active
- function: prompt-engineering, security, penetration-testing, documentation
- domain: artificial-intelligence, large-language-models, security, penetration-testing, tutorials
- platform: cross-platform
- tags: prompt-injection, jailbreak, llm-security, manual, chinese

## Member repositories
- Acmesec/PromptJailbreakManual (main) score 12

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:08:11.598291+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 2026-08-29T18:33:40.743065+00:00, confidence not recorded.
  - readme: https://github.com/Acmesec/PromptJailbreakManual (fetched 2026-08-28T04:08:11.598291+00:00, sha 4e88226301fb)
- Data as of 2026-08-30T08:39:29.467469+00:00.