r3ngine - r3ngine is the V3 rebirth of the automated reconna

项目说明

<p align="center">
<img src="frontend/public/img/banner.png" height="400px" width="520px" alt=""/>
</p>

<p align="center">
<h4 align="center"><strong>Phoenix: From the Ashes e</strong></h4>
<h3 align="center">r3ngine v3 — The Ultimate Web Reconnaissance & Scanner</h3>
</p>

<p align="center">
<h3 align="center">New V3 Dashboard</h3>
<img src=".github/screenshots/r3ngine_dash.png" height="550px" width="1020px" alt=""/>
</p>

<p align="center">
<a href=" target="_blank">
<img src=" alt="r3ngine Latest " />
</a>
&nbsp;
<a href=" target="_blank">
<img src=" alt="License" />
</a>
&nbsp;
<a href="#" target="_blank">
<img src=" alt="" />
</a>
&nbsp;
<a href=" target="_blank">
<img src=" alt="Inspired by rengine-ng" />
</a>
</p>

<h3 align="center">r3ngine 3.7.8: Hardening, CI, and Infrastructure</h3>
<p>
r3ngine v3.7.8 extends official <b>MCP Access</b> with <b>plugin-gated tools</b> — agents disco plugins via <code>r3ngine_list_plugins</code> and only then receive AD / BloodHound ingest, credential intelligence, compliance, and Burp Suite tools (sidecar <b>v1.4.0</b>). Dedicated <code>/api/mcp/</code> wrappers never open raw <code>/api/plugins/</code> routes; secrets stay redacted; Metasploit and acti remain out of MCP. Building on v3.7.6 MCP foundations (hashed keys, sessions, audit chain, singular tools, SAFE , portable cyber skills, Reacher mailbox ) and the broader Phoenix stack — <b>Target Report Generation</b>, <b>APME</b>, <b>Distributed Workers</b>, <b>Certificate &amp; Identity Intelligence</b>, <b>Exposure Correlation</b>, and Temporal-backed orchestration — this release focuses on agent-ready plugin surfaces while keeping defense-in-depth. Building off the original reNgine and further inspired by rengine-ng (Check out <a href=" target="_blank">rengine-ng v3</a> if you haven't!). Infrastructure remains <b>Django 5.2.3 LTS</b>, <b>PostgreSQL 16</b>, and <b>Gunicorn + U ASGI</b>.
</p>

!-----------------------------------------------------

<h4 align="center">Attack Path Modeling Engine</h4>
<p align="center">
<img src=".github/screenshots/apme.png" height="700px" width="1020px" alt=""/>
</p>

!-----------------------------------------------------

<h2 align="center"><a href=" target="_blank">R3NgineMobileSOC</a>: Stable Release Out Now</h2>
<p align="center">
<img src=" alt="r3ngine Mobile SOC" />
</p>

| Dashboard | Geo-Tactical Map | Scan Details | Scan Orchestration |
| :---: | :---: | :---: | :---: |
| <img src=".github/screenshots/dashboard.png" width="200px" /> | <img src=".github/screenshots/geomap.png" width="200px" /> | <img src=".github/screenshots/scan_details.png" width="200px" /> | <img src=".github/screenshots/scan_drawer.png" width="200px" /> |

The r3ngine Mobile SOC companion app provides a full command-and-control interface for managing scans, re findings, and monitoring targets from any device. Features include a 4-step scan wizard, plugin selector, real-time task log streaming, animated activity badge, ReconX monitoring settings, and geo-tactical map .

!-----------------------------------------------------

<h2 align="center"><a href=" target="_blank">r3Ngine Plugins</a>: Stable Release 1.5.0 Out Now</h2>
<p align="center">
<img src=" alt="r3ngine Plugins" />
</p>

| Acti |
| :---: |
| <img src=" width="1000px" height="450px" /> |

| Acti |
| :---: |
| <img src=" width="1000px" height="450px" /> |

| Exploit Readiness Layer |
| :---: |
| <img src=" width="1000px" height="450px" /> |

The plugin system supports dynamic installation, signed .r3n packages with Ed25519 , Temporal-wired acti, and Module Federation UI loaded directly into the host router. A plugins include Acti Intelligence, Acti, Exploit Readiness Layer, Burp Suite Integration, and Email Security.

!-----------------------------------------------------

<h2 align="center" id="r3ngine-mcp"><a href=" target="_blank">r3ngine MCP</a>: Official v1 Server Out Now</h2>
<p align="center">
<a href=" target="_blank">
<img src=" alt="r3ngine MCP v1.4.0" />
</a>
&nbsp;
<a href=" target="_blank">
<img src=" alt="Compatible with r3ngine v3.7.8+" />
</a>
&nbsp;
<a href=" target="_blank">
<img src=" alt="MCP" />
</a>
&nbsp;
<a href=" target="_blank">
<img src=" alt="TypeScript" />
</a>
</p>

The r3ngine-mcp companion (v1.4.0) is the official Model Context Protocol server for this instance. Cursor, Claude Desktop, VS Code, or any MCP-capable agent can list recon data, run a singular pipeline tool on one asset (with host-local tool_args schemas), propose follow-up batches, OSINT staging, run SAFE enrichment and attack-path proposals (propose→approve — no exploit craft), and — when plugins are installed — use plugin-gated tools for AD/BloodHound ingest, credential intelligence, compliance assessments, and Burp Suite sync. Identity, keys, sessions, and the request/response audit chain live in Settings → MCP Access. The sidecar speaks MCP only and is an HTTP client of /api/mcp/. Specialist Cursor agents: r3ngine-assessor, r3ngine-osint, r3ngine-, r3ngine-attack-path.

Install (after r3ngine is running and you ha a key):

make install-mcp / make.bat install-mcp do the same. The wrapper clones r3ngine-mcp into ./r3ngine-mcp if needed, then runs its Node setup (npm install, build, .env, live /api/mcp/ check, smoke start). Compose already wires nginx /mcp to the sidecar for HTTP transport. Full notes: documents/mcp.md and the r3ngine-mcp README.

!-----------------------------------------------------

> IMPORTANT — Upgrading from an existing installation
>
> v3.4.1 upgraded the infrastructure stack: Django 3.2 → 5.2.3 LTS, PostgreSQL 12 → 16, and the production ser from runserver to Gunicorn + U ASGI. v3.2.0 replaced Celery with Temporal. Both are breaking infrastructure changes that require a full upgrade run.
>
> You must run the full upgrade script before starting services:
>
>
>
> The script will:
> - Warn you of all changes and ask for explicit confirmation before proceeding
> - Stop and remove all existing containers
> - Back up the PostgreSQL database and upgrade it from pg12 → pg16 (automated, idempotent)
> - Rebuild all images from scratch with --no-cache
> - Apply all database migrations
> - Start the full updated stack and Gunicorn is serving
>
> Your data is safe. All Docker (scan_results, postgres_data, nuclei_templates, wordlist, etc.) are fully preserved.
>
> Do not run make up or docker compose up directly on an existing install without running fullupgrade first — migrations will not be applied automatically.
>
> Any scans running at the time of upgrade will be interrupted. Ensure no critical scans are in progress before upgrading.

!-----------------------------------------------------

Table of Contents

* About r3ngine
* Workflow
* Project Schema
* Features
* r3ngine MCP
* Quick Installation
* Administration & Recovery
* Contributing
* Reporting Security
* License

!-----------------------------------------------------

About r3ngine

r3ngine is a production-grade web reconnaissance and scanning platform. It combines a 7-tier scan pipeline with AI-powered intelligence gathering, graph-based attack path modeling, C from official threat intel feeds, and operational security controls — all orchestrated by Temporal for durable, crash-safe execution.

🦾&nbsp;&nbsp; End-to-end reconnaissance via 30+ integrated security tools: subdomain discovery, port scanning, HTTP crawling, directory fuzzing, screenshot capture, secret scanning, assessment (Nuclei, Semgrep, WPScan, Dalfox), and more.

🗃️&nbsp;&nbsp; Unified data model with a custom query language. Filter reconnaissance data using natural-language operators like http_status=200&name=admin across all finding types.

🔧&nbsp;&nbsp; Highly configurable scan engines via YAML configuration. Pre-built profiles include Full Scan, Passive Scan, Screenshot Gathering, and OSINT Scan. Every parameter — threads, timeouts, rate limits — is tunable.

💎&nbsp;&nbsp; Subscans: respond immediately to in-progress disco. Launch a targeted port scan or assessment against any subdomain without waiting for the full pipeline.

🧠&nbsp;&nbsp; C: automatic CVSS v3.1 scoring from NVD, EPSS exploitation probability from FIRST, and CISA KE — enriched on every startup and queryable via the API.

🤖&nbsp;&nbsp; Local LLM Orchestration: Run Ollama next to the stack as an opt-in compose service (COMPOSE_PROFILES=ollama) and manage its models from the LLM Toolkit dashboard to power offline risk assessments, mitigation strategy generation, and intelligent reporting without leaving your infrastructure.

📃&nbsp;&nbsp; PDF Reports: Full Scan, , and OSINT report types with customizable templates, executi, LLM-generated impact narratives, and remediation priorities.

🌐&nbsp;&nbsp; Distributed Remote Workers: Horizontally scale scanning infrastructure by deploying standalone remote workers with UI management, secure token authentication, and automatic heartbeat .

⚙️&nbsp;&nbsp; Role-based access control: Sys Admin, Penetration Tester, and Auditor roles with precisely defined permissions.

🔌&nbsp;&nbsp; MCP Access: official r3ngine-mcp v1.4.0 server so IDE agents can read recon, run singular tools with tool_args, propose follow-ups, OSINT staging, SAFE-enrich /attack paths, and use plugin-gated AD / credential / compliance / Burp tools through /api/mcp/ — hashed keys, sessions, portable cyber-skill sync, and an audit chain in Settings. The UI mirrors singular runs from the Subdomains tab (Run single tool).

!-----------------------------------------------------

Workflow

<img src=".github/workflows/workflows.png">

Temporal Scan Pipeline (v3.2.0+)

The full scan pipeline is orchestrated by MasterScanWorkflow on Temporal. Every tier boundary is a hard synchronisation point — no tier starts until all acti in the previous tier ha and their results are persisted to the database.

> (( )) = fork/join (parallel branch split/rejoin) &nbsp;·&nbsp; {{"⏸"}} = pause checkpoint (workflow waits for resume signal) &nbsp;·&nbsp; ─ requires = only runs when the noted YAML flag is set
>
> Full tier reference and execution notes: .github/workflows/temporal-scan-flow.md
>
> Project-wide code na map for contributors and AI agents: documents/PROJECT_SCHEMA.md

Project Schema

documents/PROJECT_SCHEMA.md is a living architecture map that explains ownership boundaries, workflow in, request paths, and which folders usually matter for a gi.

AI-specific onboarding entrypoints are also a in AGENTS.md, CLAUDE.md, and .cursorrules for low-token handoff into other assistants.

!-----------------------------------------------------

Features

🛰️ Distributed Workers, TLS, Identity & API Intelligence (v3.7.0)

* Distributed Remote Workers: ScanWorker model + RemoteWorkersPage Settings UI manage isolated worker pools with auth_token shared secrets and timing-attack-resistant heartbeat . Workers run headless via docker-compose.worker.yml and emit non-blocking asyncio.to_thread heartbeats; scans only route to workers acti 5 minutes.
* Certificate Intelligence: CertificateIntelligence model + tlsx ingest pipeline persist per-host issuer/subject/SAN//signature/fingerprint/wildcard/self-signed metadata. APME Certificate nodes with PROTECTS edges and x_cert_intel.yaml rules surface expired, self-signed, and weak-signature paths. /api/cert-intel/<scan_id>/ REST endpoint + CertIntelTab panel.
* Identity Infrastructure Intelligence: IdentityInfraDiscovery catalogs Okta/ADFS/Auth0/Ping/Azure AD/Keycloak/generic SAML/OIDC endpoints via a URL → title → header precedence pipeline. APME x_identity_infra.yaml rules map findings to credential-har and authenticated-access capabilities. /api/identity-infra/<scan_id>/ + IdentityInfraPanel.
* Expanded Graph & API Intelligence: APIIntelligenceProfile clusters endpoints by (base_url, api_type) with REST/GraphQL/SOAP/generic classification. New Organization (from Domain.project), Application (per webser subdomain), and APIEndpoint APME nodes with DEPENDS_ON, TRUSTS_DOMAIN, and PART_OF edges anchor the new top of chain. x_api_intelligence.yaml rules co data exfil, unauthenticated REST → account takeover, SOAP → RCE. New /api/graph/chain/ REST endpoints back a FullChainGraphTab with stat tiles and per-type drilldown. Two-pass node + edge queries with allowlist- labels replace the prior cartesian truncation.
* Exposure Correlation Engine & Taxonomy Expansion: ExposureCorrelator consolidates duplicate exposures across Nuclei, Semgrep, Trivy, and Dalfox URL + signature matching. 18 new exposure categories with MITRE technique mapping and remediation priorities, backed by 18 new tests; 14 review edge-cases hardened.
* APME Attack Trees UX: RiskSummaryBar, PriorityBadge, score weight tooltips, speculati section (showing why paths were gated out), and leaf-node detectability chips. Attack-tree 404s fixed by URL-encoding targetId. ImpactExplorer fully typed and themed; MITRE helpers consolidated into apme.utils.mitre.
* Exposure UI: ExposureList aggregate stats bar and status filter chips. ExposureCard asset summary (host + path + method). ExposureDetailsDrawer renders e as key-value list with timestamps and inline linked .
* Feroxbuster Directory Fuzzing: New run_feroxbuster engine config key runs feroxbuster alongside ffuf with shared wordlist/proxy settings; results merged and deduped on URL. Disabled by default in bundled engine YAMLs.

🔬 C & Enrichment (v3.5.0)

* C Service: Fetches CVSS v3.1 metrics from NVD API v2.0, exploitation probability scores from FIRST EPSS (synced daily feed), and syncs the CISA KEV (Known Exploited ) catalog. Enriched data is cached (7-day TTL for CVEs, 1-hour for KEV) and gracefully degrades when external APIs are una.
* History Tracking: model traces across historical scans, automatically classifying findings as new, persistent, or remediated.
* Multi-Criteria Correlation Scoring: Composite scores using configurable tool weights, asset criticality, CISA KEV/EPSS exploitability factors, and temporal modifiers. In-scan duplicate suppression groups findings under a group_key before writing.
* Neo4j ID-Based Graph Sync: C in Neo4j are linked by precise CVE ID with CVSS base score and EPSS score ingested as node properties for attack path enrichment.
* Startup Enrichment: On every orchestrator start, sync_c fires 5 minutes after graph sync completes — enriching all unenriched CVEs and refreshing the KE automatically.
* Management Command: python manage.py sync_c --all for full manual synchronization.

🧠 Intelligence & AI Hub

* Centralized AI Management: Unified interface supporting OpenAI, Anthropic (Claude 3), Google Gemini, and local Ollama models.
* Impact Intelligence: Automated impact narratives, remediation strategies, and priorities via LLMs, through interactive Cytoscape.js attack paths and a state-aware Impact Explorer.
* PII Gate Security: Ad privacy layer that anonymizes sensitive data (IPs, emails, hostnames) before sending to external LLMs.
* GPT Attack Surface Generator: Automated generation of target profiles and high-value asset identification.
* Natural Language Querying: Complex database lookups using human-like operators.

🛠️ Ad Scan & Execution Engines

* Burp Suite Professional Integration (v3.4.0 plugin): Two-phase Temporal sync (import → correlate), bidirectional scope push, filterable issues grid, and li health badge.
* Acti Intelligence Plugin: Full AD attack surface analysis — Cytoscape.js graph with 5 layout presets, real-time WebSocket streaming (150 ms batching), paginated findings API, 7-section PDF reports (ad_modern, cyber_pro templates), and RBAC e logs.
* Attack Path Modeling Engine (APME): Neo4j-based graph discovery of feasible attack routes (e.g., SQLi → DB Access → Pivot) with 179 attack rules across 13 YAML category files, MITRE ATT&CK attribution, 10-factor confidence scoring, Dijkstra pathfinding with semantic deduplication, constraint-gated (22 gates) feasibility filtering, and automated "Goal Injection".
* Adapti & Resilience Engine (ASRE): k6, wrk, hping3, and Locust orchestration with real-time ECharts telemetry Streams and WebSockets, safety kill-switches, and LLM-powered bottleneck PDF reports.
* Exploit Readiness Layer (ERL): Containerized, non-destructi with nati rotation and OpSec compliance built into the adapter layer.
* Autonomous Recon Orchestration: Temporal durable workflows with crash-safe execution, 10-attempt retry cap, full history replay at localhost:8080, and UI-based resume.
* Nmap NSE Grouped Findings: Product- grouped display in UI and PDF reports with collapsed/expandable C.
* Nuclei Sequential Se Execution: NucleiPlannerWorkflow runs se sequentially to prevent OOM on large target sets, with per-se activity status in the timeline.
* Correlation Engine: Unifies findings from Nuclei, Semgrep, Tri, Acunetix, Retire.js, and more into a prioritized threat landscape with persistent state tracking.
* Scan Queueing: Optional concurrency limiter (max 1 main + 1 subscan) with Temporal polling loop and settings panel toggle.

🕵️ Surgical Reconnaissance & OSINT

* Ad Web API Discovery: Kiterunner, Arjun, ParamSpider, LinkFinder, and InQL pipeline.
* Deep Pursuit OSINT Engine: Email pi (holehe), cross-platform social profile mapping (maigret), social presence discovery (gosearch), tactical identity permutation (username-anarchy), and a Playwright-dri Intelligence Engine with session-state + cookie-vault authentication — MFA-compatible, no stored passwords, graceful scan continuation on session expiry.
* Multi-Source Breach Intelligence: WhatBreach (DeHashed, Pastebin, Hunter.io discovery — all modules except HIBP) runs alongside the existing HIBP scraper; each EmailBreach row carries a source badge distinguishing the two tools. Optional database download is opt-in. CredSpy runs post-crawl when Microsoft Exchange/Entra infrastructure is detected (MX records, autodisco), enumerating account existence and auth capabilities GetCredentialType — proxy-mandatory to protect the operator's IP. All four report templates present breach findings grouped per email address with per-source badges and an inline CredSpy sub-table.
* URL Deduplication: Two-pass dedup after fetch_url — URL signature dedup (pre-sa parametric , content-based dedup (post-sa duplicate HTTP responses — reducing Tier 4–6 load.
* Scanning: Nuclei (sequential se, auto-template updates), Semgrep (parallel downloads, 500-file cap, 5 MB per-file limit), WPScan, Dalfox (deep scan, WAF bypass, remote payloads), CRLFuzzer, S3Scanner, Gitleaks, Retire.js.
* Custom Parameter Discovery Engine (CPDE): Define custom regex and string matchers with se levels to automatically flag undocumented or high-value parameters during crawling.
* WHOIS, WAF Detection, and IP Geolocation.

🥷 Stealth & Operational Security (OpSec)

* Enhanced Proxy Orchestration: Automated fetching, , and per-tool rotation of proxies across all discovery modules.
* Centralized Brute-Force Orchestration: Hydra and Medusa integration with Proxychains4. Multi-service targeting: SSH, FTP, HTTP, SMB, RDP, Telnet.
* OpSec Presets: User-Agent rotation, stealth timing, custom DNS resolvers, WAF bypass headers, and TOR circuit rotation.
* Hardened Scan Termination: abort_scan_history() / abort_subscan() cancels all child subscans and Temporal workflows before database updates, eliminating orphaned processes.

🎨 & Administrati

* Cyberpunk V3 UI: Glassmorphic dashboard — Hacker (Cyberpunk), Hybrid (Modern Dark), and Enterprise (Professional Slate) themes.
* Attack Surface Map v4.0: fCoSE and KLay layouts, hierarchical asset clustering (Domains > Subdomains > Endpoints), AI-dri search.
* Tactical GeoMap : CSS-animated markers and tooltip interactions for global asset positioning.
* Bounty Hub: HackerOne program management, asset tracking, and direct reporting.
* Automated Startup Sync: On every orchestrator start — graph sync, CISA KE refresh, Semgrep rule sync, stuck scan recovery, and full C — all fire as one-shot Temporal schedules.
* Configuration Export/Import: Backup and restore API keys, wordlists, tool configs, and scan engines to/from a single .zip.
* Scan Result Recovery: reco management command reconstructs the database from the scan_results on disk — idempotent, dry-run by default.
* Customizable Alerts: Notifications , Discord,

⚡ Infrastructure & Performance

* Django 5.2.3 LTS (supported until April 2028) + PostgreSQL 16 (supported until No 2028).
* Gunicorn + U ASGI: 4-worker production server with full ASGI support for HTTP and Django Channels WebSocket streams.
* Temporal Workflow Engine: Durable execution, automatic retry with configurable backoff, per-workflow cancellation.
* Automated Infrastructure Upgrade (make fullupgrade): 8-step procedure co DB backup, PostgreSQL major- upgrade (idempotent), image rebuild, migration apply, and health .
* Global Redis Caching: Unified Redis-backed caching replacing per-process local memory for shared state efficiency.
* Deterministic Resource Limits: Docker deploy.resources limits for all production services.

!-----------------------------------------------------

🛠️ De & Type Safety

The r3ngine frontend is built with a "Safety-First" philosophy under strict: true TypeScript throughout.

* Full Strict Mode: Eliminates hidden null pointers and undefined property access at build time.
* Contract Integrity: Frontend models mapped to the auto-generated OpenAPI schema (src/types/api.ts) with for tree-shaking.
* Modular Architecture: Feature-based structure — each module (targets, scans, ) maintains its own API hooks and types.
* Production Hardening: CI/CD every commit against tsc -b and vite build.

!-----------------------------------------------------

Quick Installation

Quick Setup for Ubuntu/VPS

1. Clone the repository

1. Configure the en

Ensure you change the POSTGRES_PASSWORD for security.

1. (Optional) For non-interacti, set admin credentials in .env

1. Configure Temporal worker concurrency in .env (optional)

Recommended by a RAM:

* 4 GB: TEMPORAL_MAX_CONCURRENT_ACTI=10
* 8 GB: TEMPORAL_MAX_CONCURRENT_ACTI=20
* 16 GB+: TEMPORAL_MAX_CONCURRENT_ACTI=40

The Temporal UI is a at for workflow inspection, history replay, and manual inter.

1. Run the installation script:

For non-interacti: sudo ./install.sh -n

*Note: If needed, run chmod +x install.sh first.*

r3ngine is accessible at (or your VPS IP). Do not expose port access in production.

r3ngine MCP (agent / IDE access)

1. In the UI, open Settings → MCP Access. A sys-admin sets transport to stdio, HTTP, or both, then you generate a named key. Copy the secret (r3n_mcp_…) immediately — it is shown once.

2. Install the companion server from this checkout (repo, v1.4.0):

Or: make install-mcp / make.bat install-mcp (pass extra flags via MCP_INSTALL_ARGS on Make).

3. stdio (Cursor, Claude Desktop, VS Code): the setup script can merge IDE config (--write-cursor, --write-, --write-claude). Manual snippet:

4. HTTP: when transport is http or both, Compose starts the sidecar and nginx serves with Authorization: Bearer <key>. Do not publish port 3100 on the host.

Requires Node.js 20+. See documents/mcp.md.

Installation on Other Platforms

For Mac, Windows, or other systems, refer to the installation notes in docker/ or open an issue for platform-specific guidance.

Updating

For major upgrades (including infrastructure changes), always use make fullupgrade instead of a plain docker compose up.

!-----------------------------------------------------

🔧 Administration & Recovery

Scan Result Recovery

If the database is lost or corrupted but the scan_results Docker is intact, the reco management command can reconstruct the database from files on disk.

What is recovered (when the corresponding output files exist):

| Data | Source file(s) |
|------|----------------|
| Domain | Parsed from folder name (domain_scanid) |
| ScanHistory | Folder mtime used as scan date |
| Subdomains | #id_subdomain_discovery.txt, subdomains_*.txt, subscan dirs |
| Ports + IpAddresses | #id_port_scan.txt — naabu JSONL and legacy JSON-object formats |
| EndPoints | #id_fetch_url.txt, urls_*.txt |
| | *_nmap_vulns.json, #id_nuclei_*_module.txt |
| WAF associations | #id_waf_detection.txt linked to matching subdomains |

Usage (run inside the web container):

The command is idempotent — scans already in the database are skipped on every run.

CVE Data Synchronization

C also runs automatically 5 minutes after every orchestrator startup.

Debugging

1. Enable debug mode — edit docker/web/entrypoint.sh and add export DEBUG=1 at the top, then docker-compose restart web.
2. View logs: make logs or docker compose logs temporal-python-orchestrator.
3. Temporal UI: — full workflow history, signals, event replay.
4. Disable debug mode when done: set DEBUG=0 and restart.

!-----------------------------------------------------

Screenshots

Scan Results


Live Logs


Mobile Interface


!-----------------------------------------------------

Contributing

Contributions of all sizes are welcome. Whether fixing a typo, improving UI, or adding new features — every contribution matters.

How you can contribute:
* Code impro and bug fixes
* Documentation updates
* New feature suggestions and implementations
* UI/UX enhancements
* Plugin de

To get started:
1. Check our Contributing Guide
2. Pick an open issue or propose a new one
3. Fork the repository and create your branch
4. Make your changes and submit a pull request

!-----------------------------------------------------

Submitting Issues

When submitting issues, provide as much detail as possible:

1. Enable debug mode (see Debugging above)
2. Run make logs to capture the full stack trace
3. Check the browser de console for XHR 500 errors
4. Submit a GitHub issue with the stack trace, reproduction steps, and system information

!-----------------------------------------------------

Reporting Security

Do not disclose security publicly on GitHub issues.

Go to the Security tab and click "Report a " to open GitHub's private reporting form. Include:

Steps to reproduce

Potential impact

Suggested fixes or mitigations (if any)

Reports are re within 48–72 hours. Responsible disclosure will be publicly acknowledged after the fix is released, unless you prefer to remain anonymous.

!-----------------------------------------------------

License

Distributed under the GNU GPL v3 License. See LICENSE for more information.

!-----------------------------------------------------

<p align="right"><i>Note: Parts of this README were written or refined using AI language models.</i></p>

项目信息

开发语言：Python

Star 数：12

技术要点

该项目属于安全软件方向的开源资源，代码结构清晰，适合安全研究人员学习参考。项目围绕r3ngine is the V3 rebirth of the automated reconnaissance framework for web applications with a focus on dynamic streamlined recon process, backed by a database, & simple yet intuitive User Interface. r3ngine makes it easy for penetration testers to gather recon with min config & with the help of r3ngine's correlation, making recon effortles展开，提供了可运行的实现，对理解相关安全机制、工具原理与攻防技术具有较高的研究价值。

从技术实现角度看，此类项目的核心在于对底层原理的深入理解与工程化落地。开发者在研究过程中可以结合官方文档与社区讨论，梳理项目的初始化流程、核心模块职责与数据流转，逐步掌握其中的技术细节与设计思路。

使用与学习建议

1. 下载源码压缩包，解压到本地环境。

2. 根据项目类型准备对应的运行环境与依赖。

3. 按照 README 中的步骤完成配置与部署。

4. 在隔离的测试环境中运行验证，避免影响生产系统。

免责声明

本资源来源于开源社区，仅供安全研究、技术学习与合法授权范围内的攻防演练使用，请勿用于任何非法或侵害他人权益的用途。使用本项目产生的一切后果由使用者自行承担。