{"id":"d28bc572-e97a-4014-a2c3-ae6a9cee2e30","slug":"agentburp","kind":"mcp","name":"agentburp","summary":"CLI-first, agent-native HTTP intercepting proxy & bug bounty toolkit. Headless Burp Suite alternative for AI agents (MCP), penetration testers, and ethical hackers with declarative YAML scope enforcem","owner":"jojin1709","repoUrl":"https://github.com/jojin1709/agentburp","license":"MIT","licenseTier":"permissive","trust":"community","stars":1,"match":null,"fewStars":true,"format":"mcp","createdAt":"2026-10-16T06:17:21.000Z","addedAt":"2026-10-21T06:59:03.884Z","logoUrl":"https://avatars.githubusercontent.com/u/197761551?u=ff51d671a88d4e0c5513b3a82c1e3a2c02d55ae8&v=4","linkOnly":false,"collection":null,"skillPath":null,"categories":["browser-web","developer-tools","memory-knowledge"],"tags":["ai-agents","api-security","appsec","bug-bounty","burp-suite","cli","cybersecurity","ethical-hacking"],"runtime":"go","local":true,"cloudCapable":false,"cloudCapability":"unknown","cloudReadyOnPlatform":false,"needsApiKey":false,"reviewed":false,"syncedAt":"2026-10-21T07:02:59.464Z","sample":false,"url":"https://conduits.dev/tools/agentburp","language":"Go","transports":["stdio"],"install":{},"requirements":{"env":[],"upstreamMayCharge":false},"tools":null,"readme":"> [!NOTE]\n> **Strict Scope & Authorization Requirement:** Only use `agentburp` against targets you are explicitly authorized to test. The engine refuses all traffic not listed in `scope.yml`'s `allow_hosts`.\n\n<div align=\"center\">\n\n<img src=\"./assets/banner.png\" alt=\"agentburp banner\" width=\"100%\" style=\"border-radius: 12px; margin-bottom: 20px;\"/>\n\n### A CLI-first, agent-native HTTP intercepting proxy and bug bounty toolkit.\n\nBurp Suite's core workflow — proxy, history, repeater, intruder/fuzzer, diffing, heuristic scanning, and PoC reporting — re-engineered so AI agents and terminal power-users can drive security testing entirely through JSON output and an MCP stdio server, with no heavy GUI required.\n\n[![Go Version](https://img.shields.io/badge/Go-1.22+-00ADD8?style=for-the-badge&logo=go)](https://golang.org)\n[![Go Reference](https://pkg.go.dev/badge/github.com/jojin1709/agentburp.svg)](https://pkg.go.dev/github.com/jojin1709/agentburp)\n[![Latest Release](https://img.shields.io/github/v/release/jojin1709/agentburp?style=for-the-badge&logo=github)](https://github.com/jojin1709/agentburp/releases)\n[![MCP Native](https://img.shields.io/badge/MCP-Ready-8A2BE2?style=for-the-badge&logo=anthropic)](https://modelcontextprotocol.io/)\n[![Architecture](https://img.shields.io/badge/CLI-First-black?style=for-the-badge&logo=gnubash)](https://github.com/jojin1709/agentburp)\n[![Security](https://img.shields.io/badge/Scope-Guarded-success?style=for-the-badge)](./scope.yml)\n\n**Developed by [JOJIN JOHN](https://www.linkedin.com/in/jojin-john/)**\n\n[![LinkedIn](https://img.shields.io/badge/LinkedIn-JOJIN_JOHN-0A66C2?style=for-the-badge&logo=linkedin&logoColor=white)](https://www.linkedin.com/in/jojin-john/)\n[![GitHub](https://img.shields.io/badge/GitHub-jojin1709-181717?style=for-the-badge&logo=github&logoColor=white)](https://github.com/jojin1709)\n\n<p><strong>Quick Launch</strong></p>\n\n```powershell\n# Windows (PowerShell - Works anywhere including System32)\nirm https://raw.githubusercontent.com/jojin1709/agentburp/main/install.ps1 | iex\n```\n\n```bash\n# Linux / macOS (curl)\ncurl -sSL -o agentburp https://github.com/jojin1709/agentburp/releases/latest/download/agentburp-linux-amd64 && chmod +x agentburp && ./agentburp proxy start --addr 127.0.0.1:8888\n```\n\n<sub>Zero installation required. Download and launch directly via PowerShell `irm` or `curl`.</sub>\n\n---\n\n</div>\n\n> [!TIP]\n> **AI Agents and LLMs:** Run `./agentburp mcp-serve` to connect `agentburp` directly to Claude, Cursor, or any Model Context Protocol (MCP) compatible AI agent over newline-delimited JSON-RPC stdio.\n\n---\n\n## Table of Contents\n\n- [Table of Contents](#table-of-contents)\n- [What is agentburp?](#what-is-agentburp)\n  - [Why agentburp Exists](#why-agentburp-exists)\n  - [Why \"agentburp\"?](#why-agentburp)\n  - [Designed for Agents and Human Researchers](#designed-for-agents-and-human-researchers)\n- [agentburp in Action](#agentburp-in-action)\n- [Quick Start](#quick-start)\n  - [Prerequisites](#prerequisites)\n  - [1. Build agentburp](#1-build-agentburp)\n  - [2. Configure Scope](#2-configure-scope)\n  - [3. Start the Proxy](#3-start-the-proxy)\n  - [4. Intercept, Replay, and Fuzz](#4-intercept-replay-and-fuzz)\n- [Key Capabilities](#key-capabilities)\n- [Command Reference](#command-reference)\n- [MCP / Agent Integration](#mcp--agent-integration)\n- [Architecture](#architecture)\n- [Safety Rails & Scope Guard](#safety-rails--scope-guard)\n- [Roadmap](#roadmap)\n- [Common Questions](#common-questions)\n- [Author & License](#author--license)\n\n---\n\n## What is agentburp?\n\n`agentburp` brings the core capabilities of web security testing suites into a lightweight, headless, JSON-native CLI. Built with Go and SQLite, it acts as both a high-performance terminal tool for security engineers and a native stdio tool-server for autonomous LLM security agents.\n\n<a id=\"why-agentburp-exists\"></a>\n<details>\n<summary><strong>Why agentburp Exists</strong></summary>\n\nTraditional GUI interception proxies like Burp Suite Professional and OWASP ZAP are designed for manual human clicks. When pairing with AI coding assistants, autonomous agents, and scriptable CI/CD workflows, GUIs introduce friction, heavyweight memory overhead, and clunky extension APIs.\n\n`agentburp` strips away the heavyweight UI and provides pure JSON stdout, fast SQLite persistence, strict YAML scope enforcement, and an MCP server out of the box.\n\n</details>\n\n<a id=\"why-agentburp\"></a>\n<details>\n<summary><strong>Why \"agentburp\"?</strong></summary>\n\nIt mirrors the indispensable workflow security researchers rely on in Burp Suite (Proxy -> HTTP History -> Repeater -> Intruder -> Scanner -> Report), but designed from the ground up for **Agents**: deterministic outputs, machine-parsable JSON, rate-limited execution, and MCP stdio integration.\n\n</details>\n\n<a id=\"designed-for-agents-and-human-researchers\"></a>\n<details>\n<summary><strong>Designed for Agents and Human Researchers</strong></summary>\n\nWhether you are an ethical hacker hunting bug bounties from your terminal, or configuring an autonomous AI pentester in a CI/CD pipeline, `agentburp` provides the exact tools needed to discover, isolate, and document vulnerabilities reliably.\n\n</details>\n\n---\n\n## agentburp in Action\n\nEvery operation outputs clean, structured JSON formatted for machine-consumption and human readability:\n\n| Tool Module | CLI Command | Purpose |\n|---|---|---|\n| **Proxy** | `agentburp proxy start` | Intercepts HTTP traffic and writes full request/response pairs into SQLite |\n| **History** | `agentburp hist list` / `show` | Query and filter captured traffic by host, method, status code, or body text |\n| **Repeater** | `agentburp repeat <id>` | Replay stored requests with modified headers, auth tokens, or payloads |\n| **Fuzzer** | `agentburp fuzz <id>` | Parameter fuzzing against baselines with automated anomaly detection |\n| **Diff** | `agentburp diff <a> <b>` | Structural diff of status codes, headers, and bodies between two requests |\n| **Scanner** | `agentburp scan <id>` | Heuristic candidate checks (IDOR, open redirect, CORS, JWT, SSRF hints) |\n| **Report** | `agentburp report <id>` | Generates ready-to-submit HackerOne, Bugcrowd, or Markdown PoC writeups |\n| **MCP Server** | `agentburp mcp-serve` | Exposes full toolkit to AI agents over JSON-RPC stdio protocol |\n\n---\n\n## Quick Start\n\n### 1. Download agentburp (No Git or Go Required)\n\n#### Windows: Graphical Setup Installer (NSIS — Recommended)\n- Download and run **[agentburp-setup.exe](https://github.com/jojin1709/agentburp/releases/latest/download/agentburp-setup.exe)**\n  *(Auto-configures user PATH, adds Start Menu shortcuts, and installs cleanly without administrator rights)*\n\n#### Windows: One-Line PowerShell Install\n```powershell\nirm https://raw.githubusercontent.com/jojin1709/agentburp/main/install.ps1 | iex\n```\n\n#### Linux (curl)\n```bash\ncurl -sSL -o agentburp https://github.com/jojin1709/agentburp/releases/latest/download/agentburp-linux-amd64 && chmod +x agentburp\n```\n\n#### macOS (curl)\n```bash\n# Apple Silicon (M1/M2/M3/M4)\ncurl -sSL -o agentburp https://github.com/jojin1709/agentburp/releases/latest/download/agentburp-darwin-arm64 && chmod +x agentburp\n\n# Intel Mac\ncurl -sSL -o agentburp https://github.com/jojin1709/agentburp/releases/latest/download/agentburp-darwin-amd64 && chmod +x agentburp\n```\n\n---\n\n### 2. Configure Scope (`scope.yml`)\n\n> [!WARNING]\n> By design, `agentburp` blocks **all** requests by default unless explicitly allowed in `scope.yml`.\n\nDownload the starter scope configuration:\n\n```powershell\n# Windows (PowerShell)\nirm https://raw.githubusercontent.com/jojin1709/agentburp/main/examples/scope.yml -OutFile scope.yml\n```\n\n```bash\n# Linux / macOS (curl)\ncurl -sSL -o scope.yml https://raw.githubusercontent.com/jojin1709/agentburp/main/examples/scope.yml\n```\n\nEdit `scope.yml` to specify your in-scope hosts:\n\n```yaml\nworkspace: mytarget\n\nallow_hosts:\n  - \"target.com\"\n  - \"*.target.com\"\n\ndeny_hosts:\n  - \"payments.target.com\"\n\nrate_limit_per_sec: 10\nsafe_mode: true\n```\n\n### 3. Start the Proxy\n\n```bash\n# Switch to or create a named workspace\n./agentburp workspace use mytarget\n\n# Start the intercepting proxy listener\n./agentburp proxy start --addr 127.0.0.1:8888\n```\n\nConfigure your browser, mobile device, or HTTP client to route traffic through `http://127.0.0.1:8888`.\n\n### 4. Intercept, Replay, and Fuzz\n\n```bash\n# View captured HTTP history\n./agentburp hist list --host target.com --limit 20\n\n# View request/response details for entry #12\n./agentburp hist show 12\n\n# Replay entry with an updated authorization token\n./agentburp repeat 12 --set \"Authorization: Bearer ATTACKER_TOKEN\"\n\n# Fuzz an ID parameter over numeric range 1..100\n./agentburp fuzz 12 --param id --range-from 1 --range-to 100 --confirm\n\n# Diff two requests to inspect authorization bypass or state differences\n./agentburp diff 12 13\n\n# Run heuristic vulnerability scans\n./agentburp scan 12\n\n# Generate a professional HackerOne PoC report or automated Nuclei template\n./agentburp report 12 --format hackerone\n./agentburp report 12 --format nuclei\n\n# Replay un-canonicalized raw HTTP socket bytes (smuggling / zero-day testing)\n./agentburp raw request.txt --ssl\n\n# Launch the embedded real-time web dashboard\n./agentburp ui --addr 127.0.0.1:8889\n\n# Generate an OAST collaborator token for blind SSRF / blind RCE testing\n./agentburp oast token\n```\n\n---\n\n## Key Capabilities\n\n- **CLI-First & Agent-Native**: No browser or GUI required. Works seamlessly over SSH, in scripts, CI pipelines, and with AI agents.\n- **Embedded Web Dashboard**: Optional dark-mode local UI at `http://127.0.0.1:8889` (`agentburp ui` or `agentburp proxy start --ui`) for inspecting live traffic, site maps, scanner findings, and collaborator hits.\n- **Out-of-Band (OAST) Collaborator**: Integrated receiver (`agentburp oast start|token|poll|list`) for verifying blind SSRF, blind XSS, and blind command injection callbacks.\n- **WebSocket Interception & History**: Intercepts RFC 6455 WebSocket handshakes and frames (`ws://` and `wss://`) into SQLite with direction, opcodes, and full payload logging (`agentburp ws list|show`).\n- **Expanded Heuristic Security Scanner**: Passively flags IDOR candidates, open redirects, permissive CORS, GraphQL introspection exposure, JWT `alg=none` and expiration risks, missing CSP/XCTO headers, and sensitive file disclosures (`.env`, `.git`, `actuator`, `swagger.json`).\n- **Strict Scope Guard & CLI Tester**: Match allow/deny rules with wildcard and regex support. Verify any target URL instantly with `agentburp scope test <url>`.\n- **Match & Replace Engine**: Dynamic regex and substring rewriting rules for request/response headers and bodies configured in `scope.yml`.\n- **Raw HTTP Replayer**: Send raw byte-level requests directly over TCP/TLS sockets (`agentburp raw`) without Go's HTTP library normalizing headers, spaces, or chunking.\n- **Live Traffic Stream**: Real-time 1-line transaction logging with method, URL, status code, payload size, and latency as traffic flows.\n- **Traffic Exporter & Importer (cURL & HAR 1.2)**: Export any request to an executable cURL command or export/import full session histories via standard HAR 1.2 for Burp Suite, Caido, and DevTools.\n- **Target Site Map Tree**: Aggregate discovered endpoints and query/body parameter surfaces across workspaces using `agentburp map`.\n- **Root CA Generator**: Generate and install trusted local root CAs with `agentburp proxy ca` for full HTTPS decryption.\n- **Upstream Proxy Chaining**: Chain `agentburp` traffic directly into upstream proxies like Burp Suite Pro, Caido, or corporate gateways via `--upstream <url>` or `upstream_proxy:` in `scope.yml`.\n- **Bug Bounty Header Injection**: Automatically inject researcher identifiers (e.g. `X-Bug-Bounty: handle@platform`) or auth tokens into every outgoing proxy request via `--inject-header` or `inject_headers:` in `scope.yml`.\n- **Safe Mode & Audit Logging**: Destructive fuzzing or active operations are barred without explicit confirmation (`--confirm`), and all actions are recorded in the SQLite audit log.\n- **Built-in Rate Limiting**: Global request throttling protects target servers and prevents rate-limit bans during automated scans.\n- **Zero Heavy Dependencies**: Pure Go with SQLite storage. Single binary deployment with instant startup.\n- **Automated Bug Bounty & Nuclei Reporting**: Directly outputs formatted Markdown reports structured specifically for HackerOne and Bugcrowd submissions, plus ready-to-run Nuclei YAML scanner templates.\n- **Native MCP Protocol**: Plug directly into AI agent workflows (Claude Code, Cursor, Windsurf, custom LLMs) via stdio JSON-RPC.\n\n---\n\n## Command Reference\n\n```bash\nUsage:\n  agentburp proxy start [--addr 127.0.0.1:8888] [--scope scope.yml] [--upstream URL] [--inject-header \"Name: Value\"]... [--ui] [--quiet]\n  agentburp proxy ca [--out-dir DIR]\n  agentburp ui [--addr 127.0.0.1:8889] [--scope scope.yml]\n  agentburp raw <request_file> [--ssl] [--target host:port]\n  agentburp oast start [--addr 127.0.0.1:9998] [--domain DOMAIN]\n  agentburp oast token [--server URL]\n  agentburp oast poll <token> [--server URL]\n  agentburp oast list [--server URL]\n  agentburp ws list [--entry-id N] [--limit N]\n  agentburp ws show <frame_id>\n  agentburp export [<id>] [--format curl|har] [--out FILE] [--limit N]\n  agentburp import <file.har> [--workspace W]\n  agentburp map [--host H] [--json] [--limit N]\n  agentburp scope test <host_or_url> [--scope scope.yml] [--json]\n  agentburp hist list [--host H] [--method M] [--status N] [--body S] [--limit N]\n  agentburp hist show <id>\n  agentburp repeat <id> [--set \"Header: value\"]... [--body B]\n  agentburp fuzz <baseline_id> --param NAME (--wordlist FILE | --range-from N --range-to N) [--anomalies] [--confirm]\n  agentburp diff <id_a> <id_b>\n  agentburp scan <entry_id>\n  agentburp report <entry_id> [--format hackerone|bugcrowd|nuclei|generic] [--finding-id N]\n  agentburp workspace use <name>\n  agentburp workspace current\n  agentburp mcp-serve [--scope scope.yml]\n\nGlobal:\n  --scope PATH   path to scope.yml (default \"scope.yml\"), accepted by any subcommand\n```\n\n---\n\n## MCP / Agent Integration\n\nLaunch `agentburp` as an MCP server to provide security tools directly to your LLM agent:\n\n```bash\n./agentburp mcp-serve --scope scope.yml\n```\n\nThe server communicates via newline-delimited JSON-RPC over stdin/stdout:\n\n```json\n{\"id\": 1, \"method\": \"history.search\", \"params\": {\"host\": \"target.com\", \"limit\": 10}}\n{\"id\": 2, \"method\": \"fuzz\", \"params\": {\"baseline_id\": 12, \"param\": \"id\", \"range_from\": 1, \"range_to\": 100}}\n{\"id\": 3, \"method\": \"diff\", \"params\": {\"a\": 12, \"b\": 13}}\n{\"id\": 4, \"method\": \"scan\", \"params\": {\"entry_id\": 12}}\n{\"id\": 5, \"method\": \"sitemap\", \"params\": {\"host\": \"target.com\"}}\n{\"id\": 6, \"method\": \"report\", \"params\": {\"id\": 12, \"format\": \"nuclei\"}}\n{\"id\": 7, \"method\": \"oast.token\", \"params\": {}}\n{\"id\": 8, \"method\": \"oast.poll\", \"params\": {\"token\": \"c045d02a44ad9675\"}}\n{\"id\": 9, \"method\": \"ws.list\", \"params\": {\"entry_id\": 12}}\n```\n\nWire `agentburp` into Claude Desktop, Cursor, or any MCP client by adding it to your MCP configuration:\n\n```json\n{\n  \"mcpServers\": {\n    \"agentburp\": {\n      \"command\": \"/path/to/agentburp\",\n      \"args\": [\"mcp-serve\", \"--scope\", \"/path/to/scope.yml\"]\n    }\n  }\n}\n```\n\n---\n\n## Architecture\n\n```mermaid\nflowchart TD\n    CLIENT[\"Target Traffic / Browser / Client\"] --> PROXY[\"agentburp Intercepting Proxy\"]\n    PROXY --> GUARD[\"Scope Guard & Rate Limiter<br/>(scope.yml)\"]\n    GUARD -- \"In-Scope\" --> TARGET[\"Authorized Target Host\"]\n    TARGET --> PROXY\n    PROXY -- \"Store Req/Resp\" --> DB[(\"SQLite Database<br/>~/.agentburp/history.db\")]\n\n    DB --> HIST[\"History Filter & Search\"]\n    DB --> REP[\"Repeater Engine\"]\n    DB --> FUZZ[\"Intruder / Fuzzer Engine\"]\n    DB --> DIFF[\"Diffing Engine\"]\n    DB --> SCAN[\"Heuristic Scanner\"]\n\n    SCAN --> REPORT[\"PoC Report Generator<br/>HackerOne · Bugcrowd · Markdown\"]\n    \n    REP --> MCP[\"MCP Server (JSON-RPC stdio)\"]\n    FUZZ --> MCP\n    HIST --> MCP\n    SCAN --> MCP\n    DIFF --> MCP\n    MCP --> AGENT[\"AI Agents<br/>Claude · Cursor · Autonomous Scripts\"]\n```\n\n---\n\n## Safety Rails & Scope Guard\n\n1. **Default-Deny Policy**: Nothing is proxied, repeated, fuzzed, or scanned unless the host matches an entry in `allow_hosts`.\n2. **Deny Overrides**: `deny_hosts` always overrides allow rules (useful for blacklisting sensitive payment or auth endpoints).\n3. **Safe Mode**: Enabled by default in `scope.yml`. Blocks active attacks and fuzzing unless explicitly invoked with `--confirm`.\n4. **Audit Trail**: Every active scan and fuzz operation is timestamped and recorded in the SQLite `audit_log` table with explicit reason metadata.\n5. **Rate Limiting**: Built-in token bucket ensures requests never exceed the specified `rate_limit_per_sec`.\n\n---\n\n## Roadmap\n\n- [ ] Full HTTPS MITM with auto-generated and installable CA certificates\n- [ ] GraphQL-aware query parsing and mutation fuzzing\n- [ ] Built-in OOB Collaborator (DNS/HTTP callback listener for blind SSRF/XXE)\n- [ ] Secrets and credential leak detector for response bodies & JS bundles\n- [ ] Autonomous crawler/spider for sitemaps, robots.txt, and endpoints\n- [ ] WAF & rate-limit auto-detection with exponential backoff\n\n---\n\n## Common Questions\n\n<details>\n<summary><strong>Can an AI agent run agentburp completely autonomously?</strong></summary>\n\nYes. All commands output standard JSON, and `mcp-serve` implements a stdio JSON-RPC interface designed specifically for agents like Claude or custom LLM frameworks to execute scans, analyze responses, and replay attacks without human intervention.\n</details>\n\n<details>\n<summary><strong>How does agentburp isolate different penetration tests?</strong></summary>\n\nVia workspaces (`agentburp workspace use <name>`). Each workspace segments traffic, history, findings, and audit logs so multi-target testing never cross-contaminates data.\n</details>\n\n<details>\n<summary><strong>Does agentburp require Burp Suite to be installed?</strong></summary>\n\nNo. `agentburp` is completely standalone, written in Go with an embedded SQLite engine. It has zero dependency on Burp Suite or Java.\n</details>\n\n---\n\n## Author & License\n\n**Developed with ❤️ by [JOJIN JOHN](https://www.linkedin.com/in/jojin-john/)**\n\n- **LinkedIn:** [linkedin.com/in/jojin-john](https://www.linkedin.com/in/jojin-john/)\n- **GitHub:** [@jojin1709](https://github.com/jojin1709)\n\nThis project is open-source software provided under the [MIT License](LICENSE).\n\n> **Disclaimer:** Use responsibly and only against systems you are explicitly authorized to test. The author assumes no liability for unauthorized or illegal use of this software.\n"}