- Reasoning the way each vendor expects it — adaptive thinking, effort levels, thinking budgets, and a real "off" are sent in each model family's native form.
- Three new providers — MiniMax, Mistral, and xAI, for 13 provider types in total. Any other OpenAI-compatible endpoint works through the custom provider, which now also calls Azure OpenAI…
- Tested, not just listed — every built-in provider, plus example configurations for OpenRouter, DeepInfra, Ollama Cloud, self-hosted vLLM, and others, ships with a ctester report in examples/tests/…
- Current catalogues — GPT-6, Claude Opus 5.x and Fable 5.x, Gemini 3.x, Qwen 3.8, GLM 5.3, DeepSeek V4, and Grok 4.7, with prices checked against vendor pages.
- Provider failover — LLM_FALLBACK_PROVIDER repeats a failed model call on a second configured provider and records the switch in the trace.
- Anthropic without a long-lived key — Workload Identity Federation exchanges a Kubernetes, GitHub Actions, or other OIDC token for a short-lived Claude token.
PentAGI by VXControl
A self-hosted multi-agent system that carries out penetration testing tasks in a sandboxed environment, written in Go and React and working with several model providers.
Use it in: Web app API / SDK Self-hosted Works with: Local models Own API key Multi-agent Headless / CI
On this page
Usage data: latest readings Nov 20 to Nov 21, 2026, last checked Nov 21, 06:33 UTC. Date of each number
What PentAGI does
PentAGI is an open-source autonomous penetration testing agent from VXControl, first released in January 2025. You use it through a web app, an API and self-hosted. Its documentation covers subagents or parallel agents, a non-interactive mode for scripts and CI, your own API key and local models. It can run models from OpenAI, Anthropic, Google, Amazon Bedrock, DeepSeek and Qwen and others. It is free and open source (MIT); you pay only for the models it calls.
Key facts
| Maker | VXControl |
|---|---|
| Type | Open-source autonomous penetration testing agent |
| First released | January 2025 |
| License | MIT |
| Interfaces | Web app, API, Self-hosted |
| Main language | Go |
| Open issues and PRs | 76 |
| Docker Hub pulls | 485,607 |
Pricing
Free and open source (MIT), self-hosted; users pay their own model costs.
Source: official pricing page, checked Oct 6, 2026. Prices change often; confirm before you buy.
Quick answers
- Free to start?Yes
- Open source?Yes (MIT)
- Where you use itWeb app, API, Self-hosted
- Model providersopenai, anthropic, google, amazon, deepseek, qwen, mistral, moonshot, xai, openrouter, any OpenAI-compatible, local
- Subagents / multi-agent?Yes
- Runs in scripts / CI?Yes
- Feature comparisonAll agents
- Closest alternativeARTEX · all alternatives
What's new in PentAGI: October 2026
October brought expanded model support across new providers and vendors, improved reasoning configuration for each API style, and enhanced security defaults. The release adds three new provider integrations and strengthens both authentication and container isolation.
- Three new providers—MiniMax, Mistral, and xAI—bring the total to 13 supported provider types.
- Adaptive reasoning now sends thinking budgets and effort levels in each vendor's native format rather than a generic approach.
- Provider failover automatically retries failed model calls on a second configured provider.
- Workload Identity Federation lets you use Kubernetes or GitHub Actions tokens to get short-lived Claude credentials without long-lived API keys.
1 release this month · summarised from the official release notes · all months
AgentGid articles about PentAGI
Get started with PentAGI
| Docker | docker pull vxcontrol/pentagi |
|---|
These commands use the packages tracked on this page. The vendor may recommend a different installer; see the official documentation.
PentAGI alternatives
| # | Agent | Gid Score | Downloads 7d | 7d | 30d | Stars | Latest release | Price | |||
|---|---|---|---|---|---|---|---|---|---|---|---|
| 46 |
|
502/4 | — | — | — | 14K+1,584/day | 0.6.010d ago | Free (OSS) | |||
| 87 | 41 | 12.4K | up 23.9% | up 21.1% | 76.8K+209/day | 1.7.111d ago | Free + $29/mo | ||||
| 97 |
Codex SecurityOpenAI |
392/4 | — | — | — | 14.2K+135/day | 0.10.13d ago | — | |||
| 231 |
CAI (Cybersecurity AI)Alias Robotics |
20 | 851 | up 39.1% | up 27.4% | 9,870+2/day | — | Free | |||
| — |
Dropzone AIDropzone AI |
— | — | — | — | — | — | Custom | |||
| — |
Microsoft Security Copilot agentsMicrosoft |
— | — | — | — | — | — | Usage-based | |||
Head to head: PentAGI vs ARTEX · PentAGI vs Strix · PentAGI vs Codex Security · PentAGI vs CAI (Cybersecurity AI) · PentAGI vs Dropzone AI · PentAGI vs NodeZero · PentAGI vs XBOW
Ratings from people who use PentAGI
No ratings yet. Used PentAGI? Rate it and say what it is good and bad at.
What changed: recent PentAGI releases
1 stable releases in the last 90 days · Monthly summary · Full changelog
- User Resources — a persistent, per-user file library with MD5-deduplicated storage and a virtual path filesystem.
- Flow files — per-flow workspace files that sync into worker containers at /work/uploads and /work/resources. Files can be pulled back out of a running container and promoted into the user library.
- File Manager UI — a reusable tree component with multi-select, keyboard navigation, drag-and-drop, sortable columns, bulk actions, and an overwrite workflow.
- Limits and hardening — enforced on both ends: 300 MB per file, 1000 files per request, 2 GB total, 255-byte names. Upload paths are protected against directory traversal and symlink escapes.
- GraphQL/REST CRUD plus semantic search over the knowledge store, with admin/user scoping, per-user document ownership, re-embedding on update, and real-time subscriptions.
- A new /knowledges interface with list and detail pages, a TipTap markdown editor, partial updates, inline rename/delete, and a collapsible semantic-search input (hotkey-accessible).
Earlier releases
- Token usage and cost breakdown per flow and per agent type (primary, pentester, coder, installer, searcher, adviser, etc.)
- Cache hit rates and cache read/write cost separation for Anthropic and Gemini providers
- Tool call frequency and execution time metrics per flow and subtask
- Per-model cost detail, useful when running multiple provider configurations simultaneously
- AGENT_PLANNING_STEP_ENABLED=true — enables a planning step before each specialist agent starts work, where a planner generates a 3–7 step execution plan to scope the subtask and prevent drift.
- EXECUTION_MONITOR_ENABLED=true — enables automatic detection of unproductive agent behavior: consecutive identical tool calls (EXECUTION_MONITOR_SAME_TOOL_LIMIT, default 5) and excessive…
- Gemini 2.5/3.0 family with thinking tokens support
- Anthropic Claude Sonnet 4+ with extended reasoning
- DeepSeek R1 and Kimi K2.5 in reasoning mode
- OpenAI o-series models with signature thoughts
- OpenRouter and OpenAI-compatible endpoints with reasoning content preservation
- Native caching support for Anthropic (ephemeral cache controls) and Gemini (pre-created content caching)
- Fixed Gemini provider compatibility issues preventing proper LiteLLM integration
- All providers now support LiteLLM passthrough mode with standardized endpoints
- Anthropic
- Tested and verified with LiteLLM v1.80.11-stable.1
- Enhanced Gemini provider with custom HTTP transport for API key injection and URL rewriting
- Changed file mounting scheme in PentAGI container to resolve Windows path format issues
- Added stop reason to error messages when LLM fails to generate tool calls
- If stop reason is length, increase max_tokens parameter for the affected agent in provider settings
- Improves troubleshooting and configuration optimization
- Migrated to new DuckDuckGo API with HTML response parsing
- Added comprehensive test coverage with real-world search scenarios
- Significantly improved reliability and result quality
PentAGI prices over time
PentAGI popularity and usage data
The GitHub repository has 34.7K stars, 2,993 of them added in the last 30 days. The latest stable release is 2.2.0, published Oct 5, 2026. It was mentioned in 1 Hacker News posts and comments over the last 30 days.
Download the raw daily series: pentagi.csv
Gid Score breakdown
Gid Score shows public popularity and activity (usage, GitHub stars, mentions, growth), not the quality of the product. Each component is scored 0 to 100 from public signals; a dash means no data for it. How the Gid Score works.
How fresh these numbers are
| Number | Latest reading | Last checked |
|---|---|---|
| GitHub stars | Nov 21, 2026 | Nov 21, 06:26 UTC |
| Docker pulls | Nov 21, 2026 | Nov 21, 06:27 UTC |
| Website rank (Tranco) | Nov 20, 2026 | Nov 21, 06:33 UTC |
| Hacker News mentions | Nov 20, 2026 | Nov 21, 06:27 UTC |
We fetch every source once a day. Registries publish downloads and page views a day or two late, so their latest reading trails the check. A reading more than 4 days old is marked out of date; the Gid Score still uses it until a newer one arrives.
PentAGI FAQ
How much does PentAGI cost?
Free and open source (MIT), self-hosted; users pay their own model costs.
Is PentAGI open source?
Yes. PentAGI is open source, released under the MIT license.
How popular is PentAGI?
The GitHub repository has 34.7K stars, 2,993 of them added in the last 30 days. The latest stable release is 2.2.0, published Oct 5, 2026. It was mentioned in 1 Hacker News posts and comments over the last 30 days.
What is the latest version of PentAGI?
The latest stable release we track is 2.2.0, published on Oct 5, 2026.
What are the alternatives to PentAGI?
The closest alternatives in the same category by Gid Score are ARTEX, Strix, Codex Security, CAI (Cybersecurity AI).
For the makers of PentAGI badges, corrections
Add a live badge to your README or site:
[](https://agentgid.com/agents/pentagi/)
[](https://agentgid.com/agents/pentagi/)
Something wrong or outdated on this page? Send a correction with a link to the official source. More for makers.
Where these numbers come from
GitHub: vxcontrol/pentagi. Site rank: pentagi.com on the Tranco list. See data sources for how each one is collected.
Product facts and pricing were checked against: github.com, hub.docker.com.