Security●●Solid
SentinelGate – Access control for AI agents (open-source MCP proxy)
Sub-millisecond overhead proxy blocking hallucinated tool calls before they execute.
Niche GemShip It
andreadev
602mo ago
Back to browse
GitHub Repository
Access control for AI agents. MCP proxy + Policy Decision Point. CEL policies, RBAC, full audit trail. Any container, any sandbox.
25 starsGo
SentinelGate – Universal Firewall for AI Agents (Open Source, Go)
by Sentinel-gate·Feb 18, 2026·2 points·1 comment
AI Analysis
●●●BangerSolve My ProblemBig Brain
Addresses real risk: AI agents currently run unrestricted—SentinelGate proxies all actions.
Strengths
- •Solves genuine agent safety gap: no existing tool intercepts before execution
- •Proxy sits outside agent code—works with Claude, Gemini, Python, Node with no refactor
- •CEL policies + RBAC + full audit trail covers enterprise compliance requirements
Weaknesses
- •Very early: 6 commits, no public deployments or case studies mentioned
- •Depends on MCP adoption—smaller attack surface if agents don't use MCP servers
Category
Target Audience
Developers and teams deploying AI agents that need audit trails and policy controls
Similar Projects
Security●●●Banger
AvaKill – Deterministic safety firewall for AI agents (<1ms, no ML)
Deterministic <1ms policy kill switch for AI agent tool calls, zero ML.
Solve My ProblemBig BrainShip It
duroapp
333mo ago
Security●●●Banger
Kvlar – Open-source firewall for AI agent tool calls
Fail-closed MCP security policy layer with curated templates and policy-as-code YAML.
Solve My ProblemShip It
kvlar
103mo ago
Security●●●Banger
Aegis – Open-source pre-execution firewall for AI agents
Interceptor layer blocks SQL injection and shell injection before agents execute them.
Big BrainZero to OneSolve My Problem
AEGIS_JB
203mo ago
Security●●●Banger
VellaVeto – Fail-closed runtime proxy for MCP tool calls, in Rust
Formal verification + 20 threat layers for AI agent security at sub-5ms latency.
WizardryBig BrainShip It
paolovella
113mo ago
Security●●Solid
Stop your agent running Terraform destroy or Git push origin main
Blocks terraform destroy and git push before agents execute destructive commands.
Solve My ProblemBig Brain
prudhvinomos
302mo ago