decepticon vs pentagi
Two of the top projects in Security, side by side: who leads on each score part and who each one suits, then setup, license, activity and what each review found.
decepticon
Autonomous red-team agents that run multi-stage attack chains from an approved plan
pentagi
Autonomous multi-agent penetration testing system with sandboxed security tools
Verdict
decepticon ranks higher, 84 to 79. Of the 4 score parts we measure for both, decepticon leads on 2, pentagi on 1 and 1 is even.
- Adoptionpentagi leads by 34 points, 91 to 57: a wide gap.
- FreshnessEven at 100.
- Maintenancedecepticon leads by 40 points, 100 to 60: a wide gap.
- Agent-readydecepticon leads by 30 points, 70 to 40: a wide gap.
Pick decepticon if you are
Authorized red teamers and pentesters automating engagements
- Reports 102/104 (98.08%) on XBOW validation benchmarks, with per-challenge results published
- Persistent tmux sessions let interactive tools like msfconsole and evil-winrm work
Pick pentagi if you are
Security engineers, pentesters and red teams doing authorized testing
- Runs 20+ pentest tools (nmap, Metasploit, sqlmap) in an isolated Docker sandbox
- Supports 10+ LLM providers, custom OpenAI-compatible endpoints, and local Ollama or vLLM
| What we compare | decepticon | pentagi |
|---|---|---|
| Score parts, out of 100 | ||
| Adoption | 57, popular | 91, widely used |
| Freshness | 100, active | 100, active |
| Maintenance | 100, healthy | 60, fair |
| Agent-ready | 70, partly | 40, minimal |
| Facts from GitHub and the README | ||
| Stars | 6.6k | 34.6k |
| License | Apache-2.0 (permissive) | MIT (permissive) |
| Last commit | Nov 2026 | Oct 2026 |
| Last release | Oct 2026 | Oct 2026 |
| Language | Python | Go |
| Docker | Yes | Yes |
| GPU | Not needed | Optional |
| arm64 or Apple Silicon | Mentioned | Mentioned |
decepticon
Decepticon is a red-team system with 16 specialist agents covering recon, exploitation, post-exploitation, AD, cloud and reversing. Commands run in persistent tmux sessions inside a Kali sandbox on a separate network, and findings go to a Neo4j knowledge graph. Models are routed through LiteLLM, and it ships a terminal CLI, an on-demand web dashboard and an MCP server for Claude Code and Codex.
Who it is for: Authorized red teamers and pentesters automating engagements
Strengths
- Reports 102/104 (98.08%) on XBOW validation benchmarks, with per-challenge results published
- Persistent tmux sessions let interactive tools like msfconsole and evil-winrm work
- Kali sandbox runs on its own network, separate from the management plane
- Supports ten providers including local Ollama, with tiered fallback chains
Weaknesses
- Needs Docker Compose with several services: LiteLLM, PostgreSQL, Neo4j, LangGraph, sandbox
- LangGraph drives the sandbox through the Docker socket, which widens host exposure
- Offensive Vaccine defense loop is described as planned, not shipped
- Hosted app and commercial demo are promoted; the gap to the open-source CLI is unclear
- no GPU
- Docker + Compose
- Needs Docker, Docker Compose v2, LiteLLM, PostgreSQL, Neo4j, LangGraph
- Models: Anthropic, OpenAI, Google Gemini, MiniMax, DeepSeek
pentagi
PentAGI takes a target and goal in plain language, then uses a team of specialized agents (orchestrator, researcher, developer, executor) to plan and run 20+ tools such as nmap, Metasploit and sqlmap inside an isolated Docker sandbox. Findings are stored in PostgreSQL with pgvector, with an optional Neo4j/Graphiti knowledge graph, and it produces vulnerability reports. It works with 10+ LLM providers, any OpenAI-compatible endpoint, or local models via Ollama or vLLM.
Who it is for: Security engineers, pentesters and red teams doing authorized testing
Strengths
- Runs 20+ pentest tools (nmap, Metasploit, sqlmap) in an isolated Docker sandbox
- Supports 10+ LLM providers, custom OpenAI-compatible endpoints, and local Ollama or vLLM
- Persistent memory in PostgreSQL with pgvector, optional Neo4j knowledge graph via Graphiti
- REST and GraphQL APIs with Bearer tokens; Langfuse, Grafana and Prometheus integrations
Weaknesses
- Full stack spans many containers (Postgres, Neo4j, Langfuse, ClickHouse, Grafana), heavy to self-host
- Agent profiles, plugins, MCP servers and attack-chain view are Enterprise-only
- Not a BAS or adversary emulation tool; no predefined campaigns
- Report export limited to web, Markdown and PDF; no JSON export
- GPU optional
- Docker + Compose
- Needs Docker, Docker Compose, PostgreSQL with pgvector, Neo4j (optional), Langfuse (optional), Grafana/Prometheus (optional)
- Models: OpenAI, Anthropic, Google Gemini, AWS Bedrock, Ollama
More in Security
Building with these?
We help teams pick an open-source AI stack and run it in production. Client work never changes a rank.