Leaderboard/io.github.viridis-security/injection-detector
MCP ServerScored via MCP protocol probing: initialize handshake, tools/list conformance, and ping + tool invocation performance.

io.github.viridis-security/injection-detector

github.com/viridis-security/mcp-services-sdk↗|v0.1.0

Formally-verified injection/exfiltration detector for AI agents (MCP-02).

97/100
Operational Score
Score Breakdown
Availability30/30
Conformance30/30
Performance37/40
Key Metrics
Uptime 30d
100.0%
P95 Latency
170.5ms
Conformance
Pass
Trend
What's Being Tested
Availability
HTTP health check to the service endpoint
Responded with HTTP 200 in 102ms
Conformance
MCP initialize handshake + tools/list
Valid MCP server info returned, tools/list responded
Performance
MCP ping + zero-arg tool invocation benchmarking
P95 latency: 170ms, task completion: 100%
Skills
detect_injection

Screen untrusted input for prompt/tool injection, exfiltration, and obfuscation before an agent consumes it. Returns a verdict (clean|suspicious|attack), probability, bits-at-risk (upper bound on adversarial capture per the Adversarial Landauer bound), matched canon patterns, and a recommended action (allow|sanitize|reject|escalate). Backed by Aristotle-verified theorems T-IB-02/T-IB-06/T-IB-01.

Tools
1 tool verified via live probe
verified 2d ago
Server: viridis-injection-detectorVersion: 0.1.0Protocol: 2024-11-05
Recent Probe Results
TimestampStatusLatencyConformance
Jun 10, 2026success102.7msPass
Jun 9, 2026success143.7msPass
Jun 5, 2026success170.5msPass
Jun 5, 2026success158.4msPass
Jun 4, 2026success108.8msPass
Jun 3, 2026success230.1msPass
Source Registries
mcp-registry
First Seen
May 30, 2026
Last Seen
Jun 9, 2026
Last Probed
Jun 10, 2026