14 projects
vibexray
X-ray an AI-built prototype: what is real, what is fake, what can break, and what the PM still has to decide
agentforeman
Approve, steer, and stop every Claude Code agent from one page.
awb
Benchmark harness measuring AI coding tool+workflow performance, not just model capability. 100 tasks, sigmoid scoring, 12 capability dimensions, gap analysis.
kb-arena
Compare retrieval architectures on your documentation and choose with reproducible evidence
ph-civic-data-mcp
MCP server for Philippine civic data: the full PSA OpenSTAT statistical catalog, PSGC locations, infra spending accountability, PSA economy (inflation, labor, poverty, population), one-call area profiles, earthquakes, weather, typhoons, procurement, solar/climate, air quality, vegetation
cloudwright-ai-mcp
MCP server for Cloudwright architecture intelligence
cloudwright-ai-web
Web UI for Cloudwright architecture intelligence
cloudwright-ai-cli
CLI for Cloudwright architecture intelligence
cloudwright-ai
Architecture intelligence for cloud engineers — design, cost, compare, export
power-dispatch-studio
Replay recorded Philippine WESM days, test demand and supply scenarios, and export hourly results from Python or the command line.
memory-arena
Open testbench for agent-memory evaluation with inspectable historical evidence
pip-skill
Convert any pip package into an AI coding assistant skill (Claude Code, Cursor, Windsurf, OpenCode)
aedes
A python package for PROJECT AEDES by Xavier Puspus of Cirrolytix Research Services.
parity-fairness
Measure bias from data and machine learning models.
