Security¶
The Security section is for people evaluating LegionForge against other agent platforms with a security lens. The Framework and Guardian sections cover how LegionForge enforces security. This section covers why it differs from the others and what an attacker actually faces.
Pages¶
-
STRIDE applied to agent systems. The kill-chain view: attacker goal → vector → which LegionForge layer catches it (or honestly, doesn't).
-
Side-by-side: LegionForge vs cloud agent platforms (Operator, Computer Use, Mariner) vs unguarded OSS frameworks (LangChain, AutoGen, CrewAI). Technical, not marketing.
-
What the Jan 2026 OpenClaw incident actually exposed. Which patterns LegionForge's architecture catches; which it would have caught only with operator action; which it wouldn't catch at all. Honest.
Where to go from here¶
- For the technical security model — what runs where, what the deterministic checks look like — see Framework → Security Model.
- For Guardian's specific behavior — the 7 checks, the rule storage, the canary endpoint — see Guardian → Architecture.
- For an introduction to the thinking behind the model — the LLM-is-not-trustworthy thesis, why deterministic, why sidecar — see Concepts → Security fundamentals.
Reporting vulnerabilities¶
Do not open a public issue for security vulnerabilities. Email security@legionforge.org. We respond within 5 business days. After a fix is in place and users have had a chance to update, we publish a security advisory in the affected repo with the coordinated CVE if one was assigned.