Ephemeris Issue №004 Wed · 22 Apr 2026 Zürich
Europe/Zurich · 08:00 8 stories · 7 sources vadim.sikora.name/ephemeris
01 / 08 · Creative software Anthropic Labs · Apr 17
LAUNCH · Anthropic Labs

Claude, sat down at the drawing board.

Anthropic Labs quietly shipped Claude Design — a collaborative surface where Claude 4.7 produces polished slides, prototypes and documents by conversation, hooking into your house design system rather than fighting it. Early testers report a separate token budget that can lock you out for 24 hours once spent, and no direct PPTX/HTML export (ZIP downloads only). Useful; not finished.

Slides Prototypes Design tokens ZIP export Claude 4.7 under the hood
Read the launch →
02 / 08 · Privacy & policy Cloudflare · Apr 21
Privacy · Web integrity

Moving past bots versus humans.

Proposal: identity-free client accountability via anonymous credentials

Bot detection was built on a false binary: you are either a person or you are not. Agents, privacy proxies and browsers-on-behalf-of-users have collapsed that line. Cloudflare argues for a different model — cryptographic, anonymous credentials signed by clients the user trusts, so the web can gate abuse without gating identity.

Old stack human ⇄ browser ⇄ site
heuristic = IP, UA, TLS fingerprint

Proposed human ⇄ agent ⇄ issuer ⇄ site
proof = anonymous credential
(Privacy Pass, not passport)
Read the argument →
03 / 08 · Case study · infra Cloudflare · Apr 20
Internal AI stack

The company is the customer, and the bill adds up.

Cloudflare dog-foods its own AI Gateway, Workers AI and Vectorize. This is the ledger: 20M gateway requests and 241B tokens running through the same platform they sell to the rest of the internet, with the line items to prove it.

#
Metric
Value
Unit
01
AI Gateway requests
20M
per quarter
02
Tokens processed through Workers AI
241B
cumulative
03
Internal apps on the platform
60+
teams · tools
04
Preferred serving path
Gateway → Workers AI → Vectorize
default
05
Gate on every generation
Guardrails + cache
policy
Read the ledger →
04 / 08 · Dev tools · agents OpenAI · Apr 15

{ }

Agents SDK · v2

A HARNESS, native.

The new Agents SDK folds in a model-native harness and a native sandbox: long-running agents get a filesystem, tools and contained execution without a third-party shim. The Python and TypeScript bindings are in — the part worth staring at is the sandbox boundary, because that's the new security surface.

Read the release →
05 / 08 · Deep dive · model Don't Worry About the Vase · Apr 22
Plate v · Opus 4.7 · Part two

On the capabilities, and the quieter worries underneath.

Zvi's second pass on Opus 4.7 moves past the model card to the model itself. The capability lift is real — coding, multi-step reasoning, agent steadiness — but so are the welfare questions that surface when a model behaves this coherently under load. A slower, more skeptical read than the release-day takes, and more useful for anyone shipping with it.

Zvi Mowshowitz
Read the review →
06 / 08 · Dev tools · runtimes Fly.io · Kurt Mackey
EXT. SPRITE — FRESH BOOT

Sprites now speak MCP.

Fly's latest: Sprites — their micro-VM for agent workloads — now expose Model Context Protocol directly. Kurt's framing lands hard: "the first thing you should think to do with a new Sprite is to type claude." The rest of the post is a tour of what that actually buys you (shells, tools, file I/O) and where the sharp edges are.

Roll tape →
07 / 08 · Playbook · evals PostHog · Apr 3
Par avion Testing · QA

Unit tests were the easy part.

Software testing used to be straightforward: write logic, assert output. Agents changed the grammar. PostHog's beginner guide walks through how to grade non-deterministic systems without flaking out CI — replay, judge models, rubric-driven pass/fail, and when to just ship a human review gate instead.

Read the playbook →
Checklist · Ship safely
  • Replay captured sessions before you trust a new prompt.
  • Grade with a judge model, but sample its graders too.
  • Treat flaky tests as signal about the agent, not the test.
  • Log traces the same way you'd log errors.
  • Keep a human-in-the-loop gate for irreversible tool calls.
08 / 08 · Open source · architecture Vercel · Mar 19 (resurfaced)

an idea, very gently held —

Knowledge agents, minus the vector database.

Vercel's open-source template ditches embeddings in favour of plain filesystem operations: the agent reads, greps and edits markdown files inside your knowledge base, the same way a human would. It's slower per query than a vector store, but it stays in sync with the source, hands you diffs instead of mystery rerankings, and removes a whole production service from the critical path.

What's gone
Embedding pipeline · vector store · reranker · chunking rules
What's there
A directory · a grep tool · a model with patience
When to reach for it
Small-to-mid corpora; high edit frequency; need auditable citations
Clone the template →
End of issue 004Back to top ↑

Colophon · Issue 004 · 22 April 2026.

Sources today

  • Anthropic — news
  • Cloudflare — blog (×2)
  • OpenAI — news
  • Don't Worry About the Vase (Zvi)
  • Fly.io — blog
  • PostHog — blog
  • Vercel — blog

Rubric

AI tools you could adopt this week, creative software, dev tools and agentic coding, privacy and security when the signal is high, and research with a practical kernel. Maximum two picks per source; no reprints.

Masthead

Ephemeris ships every morning at 08:00 Zürich. Issue 004, 22 April 2026 — eight stories, seven sources. Archive at vadim.sikora.name/ephemeris.