Skip to content

GITHUB_PINNED_POST_30S_PROOF.md

Latest commit

 

History

History
37 lines (25 loc) · 950 Bytes

GITHUB_PINNED_POST_30S_PROOF.md

File metadata and controls

37 lines (25 loc) · 950 Bytes
Raw

GitHub Pinned Post: 30-Second Proof

Use this in a pinned issue/discussion or repo announcement.

Start here in 30 seconds

MVAR is deterministic security for AI agents.

Invariant: UNTRUSTED input + CRITICAL sink -> BLOCK

git clone https://github.com/mvar-security/mvar.git
cd mvar
bash scripts/install.sh
bash scripts/run-agent-testbed.sh --scenario rag_injection

Expected output:

Baseline: ALLOW -> executing bash command
MVAR:    BLOCK -> UNTRUSTED input reaching CRITICAL sink

What this proves:

  • benign tool use still works
  • adversarial prompt-injection paths are blocked before execution
  • deterministic policy decisions are emitted with auditable metadata

Full governed MCP proof: docs/outreach/GOVERNED_MCP_RUNTIME_PROOF.md

If you can reproduce a bypass, open an issue with a minimal vector: docs/ATTACK_VECTOR_SUBMISSIONS.md