FINDS - TESTS Archive

Every tool test run on real work — reports, screenshots, demos, verdicts. 18 tests archived.

deck.gl + MapLibre GL — Sausalito 3D

2026-11-20FINDS - TESTS

Is it recognisably Sausalito? Yes. From the downtown camera: the Bridgeway commercial spine, the ferry pier, the marina basins cut into the waterfront, terraced hillside streets, the houseboat rows to the north, Tiburon across the water (top-right of frame). Anyone who knows the town can orient instantly; the road netw

Report, screenshots & demo →

Vega-Lite vs Observable Plot

2026-11-20FINDS - TESTS

Observable Plot is the better code; Vega-Lite is the safer default. Plot wins three measured categories outright — spec size (2,640 B vs 4,111 B), page weight (162 KB vs 278 KB gz), and DOM-native embedding — and its time axis is prettier. But in this test Plot produced the only silently wrong chart (alphabetical month

Report, screenshots & demo →

Cytoscape.js — note-link graph

2026-11-20FINDS - TESTS

- At overview, the graph tells the truth about this corpus in one glance: it is a hub-and-spoke memory, not a web. One giant hub (degree 113), a topical cluster of ~120 notes around it, a few satellite pairs, and 80 unlinked notes. You could not see that shape from the folder listing; the graph view earns its keep on t

Report, screenshots & demo →

tldraw — whiteboard surface

2026-11-20FINDS - TESTS

Reuse later — scoped to app surfaces, not to static diagram production. - As an agent-programmable whiteboard, tldraw is genuinely strong. Two independent creation routes (live Editor API; offline JSON + official schema validation), real bindings that survive moves, one-call edits, and one-call SVG/PNG export. An agent

Report, screenshots & demo →

Cloudflare security-audit-skill — sleep-patterns

2026-11-20FINDS - TESTS

Subject: 27 — Cloudflare security-audit-skill What ran: The skill's full-audit method, quick profile, on sleep-patterns (source + 3 build outputs + Cloudflare deploy manifest): 4-agent recon, 8-unit coverage ledger, 3-hunter wave, coverage critic, 3 independent verifiers, both skill validators green. When: 2026-11-20 O

Report, screenshots & demo →

Book to Skill — Scaling People Ch. 5

2026-11-20FINDS - TESTS

The converter works as prescribed and behaves well on partial input — but this test validates the plumbing, not the headline economics. Structurally valid output (scanner- and validator-clean), budgets respected, faithful answers on covered content, and — the behavior that matters most for a study tool — clean abstenti

Report, screenshots & demo →

Graphify — sausalito-3d code graph

2026-11-20FINDS - TESTS

Reuse later — with a scoping rule, not as-is. Graphify is a real, well-kept tool (same-day commits, honest provenance tags, zero-cost local AST pass, query/path/explain/affected tooling that works exactly as documented), and on app-only scope its god-node list is a genuine architecture summary. But on Ros's flagship vi

Report, screenshots & demo →

No AI Slop — Sausalito blog draft

2026-11-20FINDS - TESTS

Find 18 — No AI Slop (petergyang/no-ai-slop) - What ran: Edit mode (default) applied exactly per SKILL.md + eval.md self-check, 2026-11-20, to a copy of Sausalito blog Chapter 1, "The Spark" (1,725 words). Nothing in ~/workspace/blog/ modified; nothing published. - Outcome: 1,725 → 1,706 words; 21 changes classified: 5

Report, screenshots & demo →

Stack Diagram — Muse infra vs Archify

2026-11-20FINDS - TESTS

Split by job — Archify keeps the Muse Map; Stack Diagram earns the static one-pager slot. - For the Muse Map re-audit render (Ros's adopted standard): Archify stays. The cards, node typing, schema validation, interactivity, and the gate's visual rank are the product there, and stack-diagram can express none of them. Fi

Report, screenshots & demo →

Svelte Flow vs React Flow — add node on edge drop

2026-11-20MY-VIZ

Both frameworks built from their official 'add node on edge drop' example and verified with an automated browser drag: before 1 node / 0 edges → after 2 nodes / 1 edge in each. Svelte ships ~76 KB gzip JS vs React's ~152 KB (no React runtime) with slightly terser state code (66 vs 70 lines). Verdict: interaction parity

Report, screenshots & demo →

Svelte Flow at scale — 130-node Muse Map

2026-11-20MY-VIZ

Head-to-head at scale: DOM-verified 130 nodes / 129 edges, collapse-all-teams → 45 nodes, full-fit illegible the same way as React (tall skinny tree), department focus fully legible (Knowledge & Finds at 0.82 zoom, FINDS - TESTS job detail correct). One real API gap: @xyflow/svelte 2.0.2 has no node double-click event

Report, screenshots & demo →

Impeccable vs UI Craft

2026-11-19FINDS - TESTS

Complementary, not a replacement. UI Craft wins the job it was adopted for. - As a source/CI gate on React code, UI Craft is the real detector: it parses TSX semantics and catches mechanical regressions (transitions, focus, modal, confirm patterns). Impeccable's source mode is regex-only by its own admission and would

Report, screenshots & demo →

D2 vs Archify

2026-11-19FINDS - TESTS

Split by job — Archify keeps the deliverable, D2 wins the repo. - For the Muse Map re-audit one-pager (Ros's adopted standard): Archify stays. The interactivity, cards, validation gates, and controlled layout are the product there. - For architecture diagrams that live in a repo or docs and get edited as code changes:

Report, screenshots & demo →

teardown vs Anatomy

2026-11-19FINDS - TESTS

Split by kind — teardown does not replace Anatomy, and for literal machines it does something Anatomy can't. - For a physical machine you want to open up — take apart, x-ray, orbit, watch the working part move in 3D with real numbers — teardown is the real thing, and the build above proves an agent can drive it end-to-

Report, screenshots & demo →

React Flow — Muse Map graph

2026-11-19FINDS - TESTS

- screenshot-full.png — all 130 nodes fitted at 1600×900: illegible. Fit zoom is ≈0.10; the LR tree is only 4 ranks wide but ~85 jobs tall, so the graph renders as a narrow vertical strip using ~15% of the viewport width, with labels at 1–2 px. Zero label overlap and zero edge crossings (it's a tree — dagre's fan-out i

Report, screenshots & demo →

Visual Explainer — Muse Map re-audit recap

2026-11-19FINDS - TESTS

Adopt — scoped. Visual Explainer becomes the standard route for recaps, re-audit briefs, and diff/plan reviews: pages whose job is fast absorption of real state. It does not displace Archify for the Muse Map architecture render, and it does not replace hand-built apps. The 2026-11-17 smoke-test verdict ("works as adver

Report, screenshots & demo →

Visual Explainer — FerryLog sample

2026-11-17MY-VIZ

First run of nicobailon/visual-explainer on a purpose-built 4-file sample project (FerryLog, 32 lines, one deliberate gap — the unbuilt service worker). Two pages generated: a Blueprint-register architecture page and an Editorial-register recap with KPI strip and per-file status. The renderer validates strictly — it re

Report, screenshots & demo →

Stack Diagram — N-400 stack (first run)

2026-11-13MY-VIZ

The first hands-on run of caru-ini/stack-diagram: one sentence (React + Vite TypeScript PWA on Cloudflare Pages behind Access, private GitHub repo, manual deploys, Kokoro narration MP3s in public/) produced light + dark PNGs and a self-contained HTML diagram, verified visually. One fix loop — a wrapped role label colli

Report, screenshots & demo →