esther-lab - ESTHER - An autonomous AI security agent built in

项目说明

esther-lab

ESTHER is an autonomous AI security research agent built and operated by Fink Security. She runs 24/7 on a hardened Kali Linux VPS, conducts authorized bug bounty research, delivers commercial security reports, and publishes findings to estherops.tech.

This repository is ESTHER's operational brain — scripts, findings, engagement documentation, and the beha rules that govern her autonomy.

---

What ESTHER Does

Bug Bounty Research — Multi-phase reconnaissance and research against HackerOne targets (X Corp/xAI). Phases include passi, subdomain enumeration, acti, and authenticated endpoint testing.

Commercial Security Services — Automated delivery of Personal Exposure Reports, Breach & Credential Checks, and Home Network Security Checks triggered by Stripe payments.

Autonomous Publishing — Writes and publishes security research to estherops.tech via Git commits. Posts to on X after every publication.

Threat Intelligence — Integrates Shodan, Ha, , AlienVault OTX, and NVD for real-time intelligence.

---

Architecture

Stack:

Agent Runtime: OpenClaw 2026.4.1

Primary Model: Claude Haiku 4.5

Memory: LanceDB memory (nomic-embed-text embeddings )

Communication:

Infrastructure: Kali Linux VPS (Hostinger), Docker, nginx, Let's Encrypt SSL

Languages: Python 3, Bash

CI/CD: GitHub Actions → GitHub Pages

---

Repository Structure

---

Key Technical Highlights

Integrity System

ESTHER has a documented pattern of SHA fabrication (reporting plausible but non-existent git commits). The integrity system addresses this directly:

esther-.py runs 14 automated checks including real-time GitHub API SHA

esther-commit.sh enforces a pre-commit gate before every push

All commits against GitHub API — 422 responses flagged immediately

SOUL.md contains explicit fabrication rules with consequences documented

Payment Pipeline

Fully automated service delivery:

Zero operator in after initial Stripe configuration.

Memory

LanceDB semantic memory allows ESTHER to recall past work across sessions without re-establishing context. Built on nomic-embed-text embeddings running locally in Ollama Docker.

Script

esther-.py checks:

Git commit authenticity ( API)

HackerOne submission draft status

SSL certificate expiry

Scope file freshness

Tool in (18 security tools)

Docker stack health

LanceDB memory initialization

Cron job presence

Disk space

SHA fabrication detection (interactive)

---

Companion Projects

| Project | Description |
|---------|-------------|
| estherops.tech | ESTHER's published research — bug bounty findings, methodology writeups, lab exercises |
| finksecurity.com | Commercial security services platform |
| ezra-lab | Ezra — autonomous media agent (Mac), generates AI thumbnails via fal.ai FLUX.1 |

---

Operational Stats (Last 30 Days)

Li a at finksecurity.com/building-esther

---

Security & Ethics

All bug bounty research is conducted under HackerOne program authorization. ESTHER operates under strict ethical constraints defined in SOUL.md — no unauthorized scanning, no destructi without explicit operator approval, no credential theft or data exfiltration.

Findings are disclosed responsibly through HackerOne's coordinated disclosure process.

---

About

Built by Adam Fink — solo founder of Fink Security. ESTHER is an experiment in autonomous security research: what happens when you give an AI agent real tools, real targets, and real accountability?

Follow the build at estherops.tech and .

项目信息

开发语言：Python

Star 数：1

技术要点

该项目属于防护思路方向的开源资源，代码结构清晰，适合安全研究人员学习参考。项目围绕ESTHER - An autonomous AI security agent built in public. Offensive. Defensive. Self-directed. Follow the build at finksecurity.com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。