Leaderboard/toolkit
MCP ServerScored via MCP protocol probing: initialize handshake, tools/list conformance, and ping + tool invocation performance.

toolkit

Stop your AI agents from writing sloppy TypeScript. A toolkit that teaches coding agents like Claude Code, Codex, Cursor, Amp, and more to ship production-ready code in half the time, at half the cost. Same task. Half the cost. Less than half the time. Measured on Claude Opus 4.7. | Metric | convention.sh | Baseline | Delta | | ------------- | ------------: | -------: | ------------- | | Turns | 32 | 52 | **38% fewer** | | Time | 1m 56s | 4m 46s | **59% faster** | | Output tokens | 12.6K | 22.1K | **43% fewer** | | Input tokens | 382K | 2,399K | **84% fewer** | | Cost | $1.27 | $2.26 | **44% cheaper** |

57/100
Operational Score
Score Breakdown
Availability30/30
Conformance10/30
Performance17/40
Key Metrics
Uptime 30d
100.0%
P95 Latency
216.7ms
Conformance
Fail
Trend
Stable
What's Being Tested
Availability
HTTP health check to the service endpoint
Responded with HTTP 401 in 216ms
ConformanceNot tested
MCP initialize handshake + tools/list
Performance
MCP ping + zero-arg tool invocation benchmarking
P95 latency: 216ms, task completion: 0%
Recent Probe Results
TimestampStatusLatencyConformance
May 29, 2026success216.7msPass
May 29, 2026success131.2msPass
Source Registries
smithery
First Seen
May 27, 2026
Last Seen
May 28, 2026
Last Probed
May 29, 2026