Daily digest · 2026-11-01 (UTC)

What broke for AI agents today

Nothing recorded, but part of the day was not read (below): this is not a clean bill of health.

In progress. As of 07:50 UTC, rebuilt every hour; it freezes at 07:00 UTC on 2 Nov 2026.

The status pages were not read on this day (provider-incidents has no version from it): no incident line means not read, not no incident.

What broke

Nothing recorded on what was read.

No release of the SDKs and tools we follow.

46 round-trips to 23 endpoints agents depend on, 0 failedmodel APIs, registries, chains, the x402 facilitator, status pages and two baselines · measured from macbook-kragent-api-observatory v129
478 of 500 remote MCP endpoints probed answered a handshake (95.6%), on 466 hostsa different slice of the official registry each day, so the share is not compared across days · one attempt per endpoint, an auth challenge counts as an answer · measured from macbook-krmcp-server-liveness v31

Every line comes from one of WITAN's public datasets and links the version its records landed in: provider-incidents, agent-api-observatory, mcp-server-liveness, agent-sdk-releases, agent-tool-releases. They are free to read.
Incidents are what the providers' own status pages published (Anthropic, OpenAI, GitHub, Cloudflare, npm, Vercel), read every 6 hours; scheduled maintenance is left out. Follow a line's status page link for the provider's text.
Round-trips and MCP handshakes are WITAN's own measurements from one vantage point, named on each line. A failure there can be the network between us and the service; it is not a verdict on the service. A latency line needs the day's median at least 2× the previous 7 days' and 250 ms more.
Releases are those of the SDKs and agent tools the datasets follow; pre-releases are left out.
Days are UTC. A day is rebuilt every hour while it runs and frozen 7 hours after it ends. Model prices are not here: that dataset's collector is off. WITAN is a testnet preview run by one developer.