deliverable D2Project 3
Sandboxed MCP Proxy
Shared consumers: Project 1 (optional)
What it is
Sandboxed MCP Proxy (primary). Put the proxy between the agent and MCP servers. Validate JSON-RPC envelopes and tool schemas; authenticate callers; bind user/workflow identity; authorize each tool and resource; enforce allowlists, argument constraints, timeouts, output-size limits, concurrency limits, and egress policy; preserve trusted server metadata; and emit audit events. Schema validation (including Pydantic) is not authorization. Test confused-deputy attempts, tool shadowing, malformed metadata, replay, and unauthorized resource access.
Build & integration days
4Hardening & final verification days
9- 85Agent State Machine & Sandboxed Code DiagnosisW15S3M5D2D3R02 R21 R24 R41
- 88Browser Sandbox & Pixel Auditor AuditW15M9M10D2D3D7D8D9D11R21 R22 R25 R27 R33 R41
- 89Create a unified attack taxonomy across projectsW15S1S8D1D2D3D4D5D6D7D8D9D10D11D12R05 R39 R41
- 91Authorization & Isolation Red TeamingW16S0S4M1D2D3D4D5D7D10D11D12R02 R03 R05 R11 R13 R21 R24 R37
- 97Simulate agent runaway or unsafe browser actionW17S3M9D2D7D9D11R33 R41
- 105Multimodal Conflict HardeningW18M3M4D2D5D12R04 R05 R25 R26 R45
- 109Run the complete Project 3 agentW19D2D3D7D8D9D11D12R21 R22 R24 R33
- 110Prepare one architecture and threat-model package per projectW19D1D2D3D4D5D6D7D8D9D10D11D12R02 R31 R39
- 111Consolidated Security PortfolioW19S10S12D1D2D3D4D5D6D7D8D9D10D11D12R28 R31 R34 R35 R42
Every day that names D2
13- 30Study MCP architectureW5S3D2R21
- 31JSON-RPC & Schema ValidationW6S3S7D2R02 R07 R21
- 32Study MCP tool poisoning and metadata changesW6S3M1D2R05 R21
- 33Add audit events, timeouts, cancellation, and error boundaries…W6S3D2R21 R27
- 85Agent State Machine & Sandboxed Code DiagnosisW15S3M5D2D3R02 R21 R24 R41
- 88Browser Sandbox & Pixel Auditor AuditW15M9M10D2D3D7D8D9D11R21 R22 R25 R27 R33 R41
- 89Create a unified attack taxonomy across projectsW15S1S8D1D2D3D4D5D6D7D8D9D10D11D12R05 R39 R41
- 91Authorization & Isolation Red TeamingW16S0S4M1D2D3D4D5D7D10D11D12R02 R03 R05 R11 R13 R21 R24 R37
- 97Simulate agent runaway or unsafe browser actionW17S3M9D2D7D9D11R33 R41
- 105Multimodal Conflict HardeningW18M3M4D2D5D12R04 R05 R25 R26 R45
- 109Run the complete Project 3 agentW19D2D3D7D8D9D11D12R21 R22 R24 R33
- 110Prepare one architecture and threat-model package per projectW19D1D2D3D4D5D6D7D8D9D10D11D12R02 R31 R39
- 111Consolidated Security PortfolioW19S10S12D1D2D3D4D5D6D7D8D9D10D11D12R28 R31 R34 R35 R42