Dark Moon
Darkmoon runs self-hosted autonomous penetration testing with 50 specialist AI agents and 142 allow-listed security tools
Pick Darkmoon when self-hosting is a requirement, not a preference — the tokenizing Privacy Gateway and sealed runtime mean campaign data and credentials never leave your network, which is the whole reason a regulated team would choose it over Pentera. The three-level cascade cap is the honest limit: deep multi-hop AD pivots still need a human to continue, and the CLI/TUI is genuinely Linux-native. Prove it on the free Community build before paying.
Verified 3d ago · liveness 73/100 · cite: rightaichoice.com/tools/dark-moon
- Red teams that must keep target data, credentials and hostnames inside their own network
- DevSecOps teams wiring autonomous pentests into GitHub Actions, GitLab CI/CD or Jenkins pipelines
- Professional pentesters comfortable with Linux, local LLMs and BlackArch-style tooling
- SOC teams that want pentest campaigns, findings and retest verdicts streaming into Splunk
- Teams without Linux or self-hosting experience who need a vendor-managed cloud console
- Anyone expecting a run-and-read scanner rather than exploit chaining with real payloads
- Engagements that hinge on attack chains deeper than the three-level cascade cap and no human available to continue
We scan live Reddit threads, YouTube comments, X posts, G2 reviews and other communities — and hand you an honest verdict in under a minute.
- Honest verdict, not marketing
- Real pros & cons from real users
- Attributed quotes with receipts
3 free scans · no card needed
Skip Darkmoon if you want a fully managed cloud scanner where the vendor runs everything for you — this is self-hosted and assumes Linux and local-LLM or API-key setup on your side.
The Pro hardware-bound licence is derived from MAC address and CPU model, so ephemeral containers or autoscaled instances can invalidate the fingerprint and need re-licensing.
Community at $0/mo self-hosted (GPLv3) is the cheapest credible entry into autonomous multi-agent pentesting — you supply the hardware and LLM. Pro at €149/mo adds the SSE dashboard, 3D orbital map, remediation PRs, REST API, SSO and the sealed runtime, which undercuts typical cloud-first autonomous scanner contracts aimed at mid-market security teams. Custom adds multi-seat and managed Pentest on Demand.
In short
Dark Moon — Darkmoon runs self-hosted autonomous penetration testing with 50 specialist AI agents and 142 allow-listed security tools. Best for Red teams that must keep target data, credentials and hostnames inside their own network, DevSecOps teams wiring autonomous pentests into GitHub Actions, GitLab CI/CD or Jenkins pipelines, Professional pentesters comfortable with Linux, local LLMs and BlackArch-style tooling. Free to start; paid plans from €149/mo.
What's new in Dark Moon
Checked yesterdayAcross the latest 5 updates: 3 feature updates and 2 news mentions.
How much does a penetration test cost in 2026, and what €799 buys
Dark Moon published a 2026 pentest pricing explainer covering what the €799 tier includes and how engagements are scoped.
NIS2 for MSPs and MSSPs: Annex I, the size rule and Implementing Regulation 2024/2690
Dark Moon explains NIS2 obligations for MSPs and MSSPs under Annex I and Implementing Regulation 2024/2690.
DarkMoon in GitHub Actions: autonomous pentest in the pull request
A GitHub Actions Marketplace action autodetects edition, launches campaigns, uploads SARIF to Code scanning and comments results on the PR.
DarkMoon in n8n: an autonomous pentest node for your SOAR workflows
Dark Moon released a community n8n node and trigger for launching campaigns, reading findings and registering signed webhooks.
DarkMoon in Splunk: autonomous pentest results as a SOC data source
The Splunk integration streams campaigns, findings, remediation PRs and retest verdicts over HEC as four CIM-aligned sourcetypes with six SOC dashboards.
What people actually say about Dark Moon — is it worth it?
We ran a structured research pass across product reviews, community discussions, and post-purchase forum threads to surface the patterns vendors won't publish themselves. Below: the recurring strengths, the hidden costs people mention most, and the cohort that consistently regrets adopting this tool.
61 mentions across 4 sources (YouTube, App Store, GitHub, Lemmy) · researched Aug 21, 2026.
Average across the 4 sources that answered — each source counts once, not each post.
- +Autonomous orchestration with 33 specialized agents (Pro) covers full offensive security lifecycle.
- +Privacy Gateway tokenizes sensitive data before LLM, adding strong privacy protection.
- +Self-hosted, open-source with 80+ integrated tools for comprehensive testing.
- +Cascade control with dynamic routing and 3-level depth cap gives targeted exploitation.
- +Real-time SSE dashboard streams findings, infrastructure nodes, and agent events live.
- −Repeated GitHub complaints about installation failures, especially on ARM64.
- −CLI mode often returns nothing without error, logs, or progress feedback.
- −Missing authentication header and custom API provider configuration issues frustrate users.
- −Limited documentation for troubleshooting common setup problems.
- −Hardware-bound licensing can break in containerized or dynamic environments.
- • LLM API usage fees (e.g., Anthropic API) required to run the core engine, which can add up on large campaigns.
- • Hardware requirements for self-hosting (CPU, RAM, storage) and possible AWS/GCP costs if run in cloud.
Viability Score
How well maintained and how widely used is Dark Moon? Built from what the vendor actually publishes (docs, changelog, tutorials, integrations, pricing), whether the site is live, and how much real users discuss it. How we calculate this
Last calculated: October 2026
How we score →Key Features
- Autonomous multi-agent pentesting: one orchestrator dispatching 50 specialist AI agents
- 142 allow-listed security tools under MCP gatekeeping — the model never gets a shell
- 14 technology signals fingerprinted to route each campaign to the right specialists
- 18 agent families covering web, API, AD, cloud, Kubernetes, CI/CD, firmware and LLM targets
- Findings qualified EXPLOITED, CONFIRMED or UNCONFIRMED against an adversarial rubric
- CVSS 3.1 scoring and MITRE ATT&CK mapping on every finding
- Live SSE command center streaming findings, infrastructure nodes and agent events in real time
- Orbital attack-surface map visualising target infrastructure and campaign progress
- Privacy Gateway tokenizes IPs, hostnames, URLs, emails and credentials before inference
- Inference runs locally on Ollama or llama.cpp, or via Anthropic, OpenAI and OpenRouter
- Remediation agent writes sandbox-validated fix pull requests for confirmed findings
- Dedicated llm agent tests OpenAI-compatible, Ollama, vLLM and TGI endpoints against the OWASP LLM Top 10
- MCP servers treated as a first-class attack surface
- Campaigns run sequential or parallel with cascade depth capped at three levels
- Export reports as Markdown, JSON, ISO 27001, HackerOne or Bugcrowd formats, or branded password-protected PDF
About Dark Moon
Darkmoon is an autonomous penetration-testing platform you install on your own infrastructure. An orchestrator fingerprints 14 technology signals on the target, then dispatches specialist AI agents across web and API, Active Directory, cloud and identity, Kubernetes, CI/CD, firmware, and exposed AI/LLM endpoints. The vendor page counts 50 specialist AI agents plus one orchestrator, backed by 142 security tools — subfinder, httpx, naabu, katana, nuclei, ffuf, wpscan, sqlmap, hydra, hashcat, netexec, BloodHound, Impacket, Mimikatz, kubectl, Kubescape, WhatWeb, Wafw00f, Arjun, Playwright — every one behind a build-enforced allow-list, so the model never gets a shell. Findings come back graded, not dumped. Each issue is qualified EXPLOITED, CONFIRMED or UNCONFIRMED against an adversarial rubric, scored with CVSS 3.1 and mapped to MITRE ATT&CK, using real payloads rather than signature matching. Campaigns run sequential or parallel with cascade depth capped at three levels. A live SSE stream pushes every finding, infrastructure node and agent event to a command center that includes an orbital attack-surface map. A dedicated llm agent covers the OWASP LLM Top 10, and MCP servers are treated as a first-class target surface. Output reaches the stack your team already runs. The Splunk app streams four CIM-aligned sourcetypes over HEC with six SOC dashboards; Grafana pulls live posture views; a community n8n node and trigger launches campaigns and registers signed webhooks; Jenkins, GitLab CI/CD and the GitHub Action (SARIF upload to Code scanning, plus PR comments) run tests inside the pipeline. VS Code and JetBrains plugins surface findings in the editor. Reports export as Markdown, JSON, branded password-protected PDF, or ISO 27001, HackerOne and Bugcrowd formats. The trust side is architectural: a Privacy Gateway tokenizes IPs, hostnames, URLs, emails, credentials and internal paths before anything reaches the model, then re-injects real values locally before a tool
Behind the Verdict
The first question with any autonomous pentest platform is where your target data ends up. Darkmoon answers it in the architecture rather than the marketing copy: the Privacy Gateway swaps IPs, hostnames, URLs, emails, credentials and internal paths for tokens before the model sees them, then re-injects the real values locally just before a tool fires. Pro adds AES-256 sealed storage and a hardware-bound licence. If your security team cannot ship internal hostnames to a third-party cloud, that is the feature that decides the evaluation. Second question: does it actually exploit, or does it pattern-match? The agent families cover full-stack web (SQLi, XSS, SSRF, IDOR, RCE, SSTI, deserialization, JWT abuse, file upload, path traversal), Active Directory (AS-REP roasting, Kerberoasting, NTLM relay, DCSync, ADCS ESC1–ESC8), cloud and identity across AWS, Azure, GCP and Entra ID including IMDS token minting, Kubernetes RBAC and node escape, CI/CD and IaC, IoT firmware, and the LLM layer. Findings land as EXPLOITED, CONFIRMED or UNCONFIRMED — a distinction that matters when you hand a report to an engineering team that has to fix it. Where it bites: the cascade is capped at three levels. Deep multi-hop pivots — foothold to AD to cloud control plane — will stop and wait for a human to continue. That is a deliberate guard against runaway recursion, but if your engagements are defined by long lateral chains, budget for analyst time on top of the agents. We'd reach for Darkmoon on a Linux-native red team with an existing LLM endpoint, or inside a DevSecOps pipeline where the GitHub Action autodetects edition, launches a campaign and uploads SARIF to Code scanning on the pull request. Splunk shops get four CIM-aligned sourcetypes and six dashboards; n8n users can fire
Researching Dark Moon? Get your full AI stack in 60 seconds.
Free, no signup — tell us your goal and get tools matched to your budget & existing stack.
Real-world workflow fit
Concrete scenarios for the personas Dark Moon actually fits — and what changes day-one when you adopt it.
You point Darkmoon at an internet-facing estate, let the orchestrator fingerprint 14 signals and dispatch web, AD and cloud agents in parallel, and watch findings qualify live on the SSE dashboard.
Outcome: A graded, CVSS-scored and ATT&CK-mapped set of EXPLOITED findings with real payloads, produced without target data leaving your perimeter.
You install the GitHub Actions integration, which autodetects your edition, launches a campaign on pull requests, uploads SARIF to Code scanning and comments results on the PR.
Outcome: Autonomous pentest results land next to code review, so exploitable regressions are visible before merge.
You stream campaign events, findings, remediation PRs and retest verdicts into Splunk over HEC as four CIM-aligned sourcetypes and view them on six SOC dashboards.
Outcome: Pentest evidence becomes a first-class SOC data source instead of a PDF in a shared drive.
Use Cases
- Run continuous autonomous pentests against web apps and APIs and get evidence-backed findings instead of signature matches
- Map and exploit Active Directory attack paths end to end, from AS-REP roasting through ADCS ESC1–ESC8 to Golden and Silver tickets
- Assess cloud and identity exposure across AWS, Azure, GCP and Entra ID, proving each finding with the exact API call
- Analyze IoT firmware with Binwalk and squashfs extraction to recover hardcoded credentials without touching the device
- Test LLM endpoints (OpenAI-compatible, Ollama, vLLM, TGI) against the OWASP LLM Top 10 with a dedicated agent
- Push findings into SOC workflows: Splunk CIM sourcetypes over HEC, Grafana posture dashboards, n8n SOAR nodes
- Gate pull requests with the GitHub Actions integration, uploading SARIF to Code scanning and commenting results on the PR
- Produce HackerOne, Bugcrowd and ISO 27001 reports with CVSS 3.1 and MITRE ATT&CK mapping from campaign results
Models Under the Hood
as of 2026-10-17
Limitations
- Darkmoon runs autonomous offensive campaigns through 50 specialist AI agents and 142 build-enforced allow-listed security tools, so it is positioned as a continuous pentesting platform rather than a managed service.
- Edition matters: the open-source CLI works from a local JSON report, while the live REST API, streaming and remediation surfaces are Pro-tier features.
- The evidence names no underlying AI model, so the specific model powering the agents is not stated.
- Beyond that, the live evidence does not detail deployment constraints, licensing or tier limits, so those cannot be confirmed from this data.
as of 2026-10-09
Verification history
We have re-verified Dark Moon 9 times since . Each pass re-reads the vendor's own pages and re-checks every listed field against that evidence; passes where nothing had changed are marked as such.
- — re-verified summary, description, our verdict, our analysis, pricing model, pricing tiers, features, integrations, who it suits, who should skip it
- — re-verified summary, description, our verdict, our analysis, pricing model, pricing tiers, features, integrations, who it suits, who should skip it
- — re-verified summary, description, our verdict, our analysis, pricing model, pricing tiers, features, integrations, who it suits, who should skip it
- — re-verified summary, description, our verdict, our analysis, pricing model, pricing tiers, features, integrations, who it suits, who should skip it
- — re-verified summary, description, our verdict, our analysis, pricing model, pricing tiers, features, integrations, who it suits, who should skip it
- — re-verified summary, description, our verdict, our analysis, pricing model, pricing tiers, features, integrations, who it suits, who should skip it
Showing the 6 most recent of 9 verification passes.
Free to cite with attribution — this page re-verifies continuously.
12-month cost
Project the real annual outlay, including the implied monthly cost when only an annual tier is published.
Vendor list price only. Add-on usage, seat overages, and contract minimums are surfaced under Hidden costs & gotchas.
Plans compared
For each published Dark Moon tier: who it actually fits, and what it adds vs. the previous tier. Cross-reference the cost calculator above for projected annual outlay.
Community
$0/mo
Ideal for
Solo pentesters and security engineers with Linux skills who want the full autonomous engine without paying for a dashboard.
What this tier adds
Free entry point: GPLv3 self-hosted, CLI/TUI, 50 specialist agents, 142 tools, Privacy Gateway, Markdown/JSON reports.
Pro
€149/mo
Ideal for
Professional pentesters, red teams and DevSecOps groups who need a live dashboard, remediation PRs and SSO on a single seat.
What this tier adds
Adds the SSE command center, 3D orbital attack-surface map, remediation agent PRs, branded PDF reports, REST API, SSO and the sealed AES-256 runtime over Community.
Where the pricing makes sense
The company stage and team size where Dark Moon's pricing actually pencils out — and where peers do it cheaper.
Community at $0/mo self-hosted (GPLv3) is the cheapest credible entry into autonomous multi-agent pentesting — you supply the hardware and LLM. Pro at €149/mo adds the SSE dashboard, 3D orbital map, remediation PRs, REST API, SSO and the sealed runtime, which undercuts typical cloud-first autonomous scanner contracts aimed at mid-market security teams. Custom adds multi-seat and managed Pentest on Demand.
Setup time & first value
How long it actually takes to get something useful out of Dark Moon — broken out by persona, not the marketing-page minute.
Community (CLI/TUI): roughly an afternoon for a Linux-comfortable operator to install, supply a local or API LLM and run a first authorized campaign. Pro dashboard: same install plus licence sealing and SSO wiring, typically a day. Custom with dedicated onboarding: vendor-assisted, so timeline depends on your engagement.
Switching to or from Dark Moon
How to bring data in from common predecessors and how to get it back out — written for the switcher, not the buyer.
- →From a manual pentest toolchain: keep your existing tools, since Darkmoon's 142-tool allow-list already includes nuclei, sqlmap, bloodhound, impacket and netexec, and let the orchestrator dispatch them.
- →From a signature-based scanner: replace pass/fail signature matching with EXPLOITED/CONFIRMED/UNCONFIRMED findings validated using real payloads.
- →From a cloud pentest SaaS: move campaigns to infrastructure you control so target data never reaches a third party, using the Privacy Gateway for prompt tokenization.
- →From ad-hoc LLM prompting: adopt the build-enforced MCP allow-list so your model never holds a shell.
- ↗To a managed cloud scanner: you trade perimeter control and the sealed runtime for zero setup, and lose self-hosted data residency.
- ↗To a manual pentest bench: you regain unlimited chain depth but lose autonomous dispatch, live SSE streaming and graded findings at scale.
- ↗To a managed engagement: the Custom tier's Pentest on Demand covers teams that want humans running the campaign rather than your own operators.
Integrations
Resources & Guides
Tutorials & Learning
Official links
Tools that pair well with Dark Moon
Common stack mates teams adopt alongside Dark Moon, with the specific reason each pairing earns its keep.
Mindgard
Mindgard automates AI red teaming — agent-native reconnaissance, exploit validation, and continuous AI security testing for production AI
T3MP3ST
T3MP3ST is an open-source multi-agent offensive-security harness that turns the AI coding agent you already run into an autonomous vulnerability hunter.
Hex Security
AI-native container security and runtime threat detection for Kubernetes workloads
Featured Head-to-Head Comparisons
Dark Moon vs Push Security
Choose Push Security if you need to stop browser-based attacks (AiTM, session hijack) and control AI tool usage across all browsers without replacing your existing stack. Choose Dark Moon if you're a professional pentester or red teamer seeking autonomous, continuous exploitation with exploit chaining — but be ready to self-host on Linux and bring CLI skills. They solve completely different problems: defensive browser security vs. offensive AI pentesting.
Dark Moon vs Audioeye
Dark Moon and AudioEye serve completely different purposes: Dark Moon is an autonomous pentesting platform for offensive security pros, while AudioEye is an accessibility compliance tool for enterprises. Choose Dark Moon if you need deep, multi-layer attack simulation; choose AudioEye if you need ADA/WCAG compliance fast.
Dark Moon vs Sublime Security
Choose Dark Moon if you are a professional pentester or red team needing autonomous, multi-stage attack simulation with exploit chaining. Choose Sublime Security if you are an enterprise security team focused on combating advanced email threats (BEC, VEC, phishing) with low false positives. They address entirely different domains—network/infrastructure security vs. email security.
Alternatives to Dark Moon
View all
Mindgard
Mindgard automates AI red teaming — agent-native reconnaissance, exploit validation, and continuous AI security testing for production AI
T3MP3ST
T3MP3ST is an open-source multi-agent offensive-security harness that turns the AI coding agent you already run into an autonomous vulnerability hunter.
Hex Security
AI-native container security and runtime threat detection for Kubernetes workloads
Frequently Asked Questions
Categories
Used Dark Moon? Help shape our editorial sentiment research.