Public developer tools and reproducible research instruments for software, data, and intelligent systems.
| Instrument | What it reveals | Start |
|---|---|---|
| piiscope | Personal data and privacy risk across files and databases | GitHub · pip install piiscope |
| release-intel-mcp | Release context across commits, pull requests, issues, and contributors | GitHub · npx -y @barissozudogru/release-intel-mcp |
| test-intel-mcp | Coverage gaps, likely test mappings, and test priority signals | GitHub · npx -y @barissozudogru/test-intel-mcp |
| envdrift | Missing keys, type mismatches, and placeholders across environment files | GitHub · npx @barissozudogru/envdrift |
| healthcheck-gen | A reviewable Docker health check inferred from an existing Dockerfile | GitHub · npx @barissozudogru/healthcheck-gen |
| portscan-dev | The local processes listening on development and service ports | GitHub · npx @barissozudogru/portscan-dev |
| dep-health | Dependency freshness, maintenance, deprecation, and adoption signals | GitHub · npx @barissozudogru/dep-health |
Each instrument has a focused page with capabilities, use cases, limits, package links, and public source where available.
Five live model worlds share one experimental grammar: perturb a system, run seeded replicates, and measure what changed.
| World | System under observation |
|---|---|
| origin | Evolution and artificial life |
| swarm | Collective motion and local rules |
| morph | Reaction diffusion and pattern formation |
| market | Agent trading and price dynamics |
| social | Opinion networks and polarization |
petri-bench tests whether model agents can investigate these systems through controlled experiments. The same experiment layer is available to MCP clients through petri-labs-mcp.
