marq

<div align="center"> <img src="docs/logo.svg" alt="marq" width="320"> ### Your universal cyber assistant Ask anything across security and get the right tools and current expertise to answer it — on whatever model you choose, local or frontier. [](https://github.com/rhaist/marq/actions/workflows/ci.yml) </div> ```text "Threat-model this payment API." → architecture "Map our pentest findings to ISO 27001 and SOC 2." → GRC + standards "What does this malware sample actually do?" → malware "Run an authorized pentest of *.acme.com." → offensive "Our breach-notification clock under GDPR + US law?" → regulation "Build the ransomware incident runbook." → IR + resilience ``` **~80 security tools + ~73 expert playbooks (current 2026 standards), in one auditable binary.** You bring the model; marq brings the tools and the knowledge. ## Start ```bash # Get marq (prebuilt image — or build from source: docker build -t marq .) docker pull ghcr.io/rhaist/marq && docker tag ghcr.io/rhaist/marq marq ``` No Docker Desktop needed — **OrbStack** (macOS) or **Podman / Colima** (Linux) run the image just as well ([`docs/SETUP.md`](docs/SETUP.md)). ### Run it fully local — recommended Your own uncensored model, on your box: no cloud, no refusals, nothing leaves the host. Drive marq from [Pi](https://pi.dev/) over [llama.cpp](https://github.com/ggml-org/llama.cpp): ```bash # 1. the model — full flags matter: --reasoning-format keeps chain-of-thought # out of the reply, --jinja makes tool calls parse (see docs/SETUP.md §2a) llama-server -hf HauhauCS/Gemma4-12B-QAT-Uncensored-HauhauCS-Balanced:Q4_K_M \ --host 127.0.0.1 --port 8080 -ngl 99 --ctx-size 65536 --jinja \ --reasoning-format deepseek \ -fa on -ctk q8_0 -ctv q8_0 \ --temp 0.6 --top-p 0.9 --top-k 64 --min-p 0.05 --repeat-penalty 1.1 # 2. marq in a long-lived container bound to your workspace install -m 0755 pi/marq ~/.local/bin/marq && marq up ~/work # 3. one-time Pi config — llama.cpp provider + marq's system prompt + skill. # Without SYSTEM.md a small model narrates instead of driving marq. → docs/SETUP.md §4 ln -sf "$PWD/pi/SYSTEM.md" ~/.pi/agent/SYSTEM.md # then run `pi` ``` Then just talk to it: _"Set scope to scanme.nmap.org and run a quick nmap,"_ or _"Load the nis2-dora skill — what's our breach clock?"_ Full local walkthrough (Pi provider, skill, tuning) → [`docs/SETUP.md`](docs/SETUP.md#4-run-with-a-local-model-pi--the-marq-skill). ### Or fall back to a frontier model For the heaviest reasoning (architecture, GRC, IR write-ups), point a frontier MCP client at marq instead: ```bash claude mcp add marq -- docker run --rm -i \ --security-opt no-new-privileges:true --cap-drop ALL \ --cap-add NET_RAW --cap-add NET_ADMIN --cap-add NET_BIND_SERVICE \ -v marq-audit:/var/log/marq -v marq-work:/work -e MARQ_OPERATOR=marq marq ``` Any MCP client works (Codex, Claude Desktop) → [`docs/CLIENTS.md`](docs/CLIENTS.md). (The caps are only what SYN scans need; full config incl. API keys in [`mcp.json.example`](mcp.json.example).) > **Authorized use.** Advisory/knowledge is open; scanning & exploitation are > authorized-only and audit-logged — [`docs/SECURITY.md`](docs/SECURITY.md). --- ## What you get - **Execution** — ~80 wrapped Kali tools (recon, web, AD/internal, exploitation, cred cracking, malware static analysis, OSINT), each audit-logged and sandboxed; findings render to `findings.md` / `.csv` (with optional CVSS + CWE). - **Knowledge** — ~73 `load_skill` playbooks across 14 domains, web-researched to current (2026) standards, telling the model _how_ to use the tools and _what_ to do where there's no tool: offensive, malware, threat-intel, sec-ops, architecture, GRC, standards & EU/US regulation, CISO, red/purple, resilience, human factors, DevSecOps/privacy. - **Auditable by design** — every external-tool call funnels through one choke point: an append-only, `fsync`'d audit log, a timeout, and an output cap. File access is sandboxed to `/work`+`/tmp`, and the image runs non-root (the recommended run args also drop all Linux capabilities + `no-new-privileges`). <details> <summary><b>Full tool suite</b> — ~80 tools on a Kali base (click to expand)</summary> | Category | Tools | | :------------------------------------ | :------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | | **Recon / network** | `nmap` · `masscan` · `naabu` · `dnsx` · `dnsrecon` · `subfinder` · `httpx_probe` · `dns_lookup` · `whois_lookup` · `ssh_audit` · `fping_sweep` · `snmp_walk` · `snmp_check` · `snmp_brute` · `smtp_user_enum` · `smtp_test` · `asnmap` · `cdncheck` · `censys_search` | | **OSINT — org/domain** | `theharvester` · `spiderfoot` · `shodan_host` · `shodan_search` · `gitleaks` · `trufflehog` · `gau_urls` · `exif_metadata` | | **OSINT — people** | `sherlock` · `maigret_username` · `holehe_email` · `h8mail_breach` · `phoneinfoga` | | **Web app** | `nuclei` · `nikto` · `feroxbuster` · `katana` · `ffuf` · `gobuster_dir` · `arjun` · `whatweb` · `wafw00f` · `cmseek` · `wpscan` · `testssl` · `dalfox` · `sqlmap` · `jwt_tool` · `trivy` · `interactsh` · `paramspider` · `sstimap` | | **Exploitation** | `donut` · `hydra` · `searchsploit` | | **AD / internal network** | `impacket_secretsdump` · `impacket_kerberoast` · `impacket_asreproast` · `impacket_psexec` · `impacket_wmiexec` · `impacket_ntlmrelayx` · `netexec` · `certipy_find` · `bloodhound_collect` · `evil_winrm` · `enum4linux` · `smb_enum` · `ldap_search` · `responder` · `nbtscan` | | **Credentials** | `john` · `hashcat` · `hash_identify` | | **Malware research** (static) | `capa` · `yara_scan` · `olevba` · `bin_headers` | | **Files** (sandboxed `/work`, `/tmp`) | `list_dir` · `read_file` · `write_file` | | **Findings & jobs** | `report_finding` · `render_report` · `list_jobs` · `job_status` | | **Knowledge** | `load_skill` | | **Escape hatch** (opt-in) | `run_shell` | </details> <details> <summary><b>How it works</b> — one registry, two front-ends (click to expand)</summary> ``` MCP client ──stdio JSON-RPC──▶ marq serve ┐ terminal agent (Pi) ──bash──▶ pi/marq shim ─┤ both hit the same registry (docker exec) │ ▼ registry.All() ── one Tool list ├─ recon nmap, naabu, dnsx, masscan… ├─ osint theHarvester, spiderfoot, shodan, gitleaks… ├─ people sherlock, maigret, holehe, phoneinfoga… ├─ web nuclei, katana, feroxbuster, sqlmap… ├─ exploit donut, hydra, searchsploit ├─ internal impacket, netexec, certipy, bloodhound, enum4linux… ├─ creds john, hashcat, name-that-hash ├─ malware capa, yara_scan, olevba, bin_headers ├─ files read/write/list (sandboxed /work) └─ extras report_finding, render_report, list_jobs, load_skill │ runner.Run ──▶ audit.jsonl (every call) load_skill ──▶ skills library (markdown, ~73 playbooks / 14 domains) ``` Every exec tool funnels through `internal/runner/runner.go::Run` — the single point for audit logging, timeouts, and output truncation. </details> --- ## Documentation | Doc | What's in it | | :------------------------------------- | :------------------------------------------------------- | | [`docs/SETUP.md`](docs/SETUP.md) | Per-OS install (macOS & Debian), both run modes | | [`docs/CLIENTS.md`](docs/CLIENTS.md) | Pick a client/model — Claude Code, Codex, Pi + llama.cpp | | [`docs/USAGE.md`](docs/USAGE.md) | Every tool, env vars, API keys, reading the audit log | | [`docs/SECURITY.md`](docs/SECURITY.md) | Legal/ethical baseline, the guardrail model, hardening | ## License **GNU AGPLv3** — see [`LICENSE`](LICENSE). © 2026 marq contributors. Free software; run a modified version as a network service and you must offer users its source. For authorized security testing and education, **no warranty** — use responsibly.