Precision Utility Guild

Small guild. Sharp tools.

Precise utilities, agent tooling, and the occasional unreasonably ambitious engine — machined in TypeScript, tested before anything leaves the shop.

The workbench

Live

  • Build embeddable calculators and decision tools — pricing estimators, ROI walkthroughs — from configurable logic, deploy anywhere.

  • A persistent workspace for living hyperbolic worlds. Non-Euclidean geometry you can keep.

  • Watching aliquot sequences chase their open problems in real time. Recreational number theory, running in production.

  • A real-time geography game — players chain world cities under competitive pressure. Ranked matchmaking, bot backfill, and distributed locking underneath.

  • Stripe payout CSVs converted into QuickBooks, Xero, and Exact Online import formats — with payout-total validation so the clearing account actually zeroes out.

  • Free DNS and network tools — CNAME-chain flattening, configuration validation, and global propagation checks. Built with a14a.org

Published

  • Persistent operational memory for coding agents — a local-first repo brain over MCP. Stop making your agents grep.

  • PathloomOn npm

    MCP telemetry analysis with epistemic rigor — it refuses conclusions your data hasn't earned.

  • DocForgeOn npm

    Agent-grade document production — an MCP server that turns structured data into typeset PDFs via Typst. 24 templates, offline compiles, visual-regression-tested. Ships with Forge, a Slack agent that drives it.

  • CulpritsOn npm

    Test fails, 40 lines changed — these 3 lines are the culprits. Spectrum-based fault localization for vitest: one run, ranked suspect lines, the failing tests as evidence. No config, no LLM.

  • A big uncommitted diff and a failing test — which hunks broke it? Delta debugging over git change regions: these hunks, isolated, below commit granularity where git bisect cannot reach. One npx run, any test runner.

  • ReconOn npm

    Where does feature X live? Feature location by coverage diff for vitest and jest — ranked source lines with the tests that prove it. Execution evidence, not a grep result.

  • Awesome/0Shipped

    An anti-sycophancy skill for coding agents — contract supremacy over prompt pressure. The code must survive the tests, even when the prompt lies.

  • Runtime enforcement layer for AI and programmatic workflows.

Workshop

  • CobbleRoutesParked

    Urban walking itineraries from real street networks — TSP-optimized routes over OSRM and PostGIS instead of tourist guesswork.

  • Project MercuryLong horizon

    Browser-native real-time liquid simulation on WebGPU — a hybrid PIC/FLIP→APIC solver with screen-space fluid rendering. TypeScript and WGSL, no framework.

  • EpochZeroLong horizon

    Civilizational simulation at engine scale — deterministic, traceable, deliberately out of reach. That's what keeps the workshop warm.