Daily digest · 2026-11-06 (UTC)
What broke for AI agents on 6 Nov 2026
6 incidents on 3 status pages.
Final. Built 2026-11-07 07:05 UTC, 7 hours after the day ended; this page will not change.
What broke
- Cloudflare: “Fedramp Customers have Delayed Purges”minor impact · opened 17:47 UTC · identified as of Nov 7 00:00 UTC · CDN Cache Purgestatus page ↗provider-incidents v97
- Cloudflare: “Increased Errors for Durable Objects in Western Europe”minor impact · opened Nov 5 13:23 UTC · resolved 09:05 UTC (19 h 41 min) · Durable Objectsstatus page ↗provider-incidents v96
- Cloudflare: “Network Performance Issues in Singapore”minor impact · opened 08:54 UTC · resolved 10:43 UTC (110 minutes) · Singapore, Singapore - (SIN)status page ↗provider-incidents v96
- Cloudflare: “Wrangler Deployments, Pages, and Workers Builds failures”minor impact · opened 09:47 UTC · resolved 10:08 UTC (21 minutes) · Pages | Workers Builds | Wranglerstatus page ↗provider-incidents v96
- GitHub: “Degraded performance for multiple GitHub services”minor impact · opened Nov 5 23:36 UTC · resolved 00:20 UTC (45 minutes) · Actions | Git Operations | Issues | Pagesstatus page ↗provider-incidents v95
- OpenAI: “ChatGPT Work and Dots Errors”the feed gives no impact level · first seen by our reader Nov 5 18:19 UTC (the feed gives no start time) · resolved 04:31 UTC · ChatGPT Work | Codex Web | dotsstatus page ↗provider-incidents v95
Shipped
- @google/genai 3.0.0 (npm) — a new major versionreleased 04:34 UTC · previous 2.32.0agent-sdk-releases v112
- google-genai 3.0.0 (pypi) — a new major versionreleased 02:29 UTC · previous 2.34.0agent-sdk-releases v112
20 other releases on the day
- @anthropic-ai/claude-agent-sdk 0.3.319 (npm)released 00:59 UTC · previous 0.3.318agent-sdk-releases v114
- ai 7.0.164 (npm)released 19:44 UTC · previous 7.0.163agent-sdk-releases v115
- anthropics/claude-agent-sdk-python v0.2.179published 17:57 UTC · previous v0.2.178release notes ↗agent-tool-releases v114
- anthropics/claude-code v2.1.319published 17:04 UTC · previous v2.1.318release notes ↗agent-tool-releases v114
- BerriAI/litellm v1.108.2published 22:07 UTC · previous v1.108.1release notes ↗agent-tool-releases v115
- claude-agent-sdk 0.2.179 (pypi)released 17:56 UTC · previous 0.2.178agent-sdk-releases v114
- crewAIInc/crewAI 1.15.36published 04:40 UTC · previous 1.15.35release notes ↗agent-tool-releases v112
- langchain-ai/langgraph cli==0.4.42published 03:04 UTC · previous cli==0.4.41release notes ↗agent-tool-releases v112
- langchain-ai/langgraph sdk==0.4.14published 03:03 UTC · previous sdk==0.4.13release notes ↗agent-tool-releases v112
- litellm 1.108.2 (pypi)released 22:03 UTC · previous 1.108.1agent-sdk-releases v115
- openai 7.53.0 (npm)released 09:20 UTC · previous 7.52.0agent-sdk-releases v113
- openai/openai-node v7.53.0published 09:06 UTC · previous v7.52.0release notes ↗agent-tool-releases v113
- vercel/ai @ai-sdk/[email protected]published 19:46 UTC · previous @ai-sdk/[email protected]release notes ↗agent-tool-releases v114
- vercel/ai @ai-sdk/[email protected]published 19:46 UTC · previous @ai-sdk/[email protected]release notes ↗agent-tool-releases v114
- vercel/ai @ai-sdk/[email protected]published 19:46 UTC · previous @ai-sdk/[email protected]release notes ↗agent-tool-releases v114
- vercel/ai @ai-sdk/[email protected]published 22:23 UTC · previous @ai-sdk/[email protected]release notes ↗agent-tool-releases v115
- vercel/ai @ai-sdk/[email protected]published 22:23 UTC · previous @ai-sdk/[email protected]release notes ↗agent-tool-releases v115
- vercel/ai @ai-sdk/[email protected]published 19:46 UTC · previous @ai-sdk/[email protected]release notes ↗agent-tool-releases v114
- vercel/ai [email protected]published 22:21 UTC · previous [email protected]release notes ↗agent-tool-releases v115
- vercel/ai [email protected]published 19:43 UTC · previous [email protected]release notes ↗agent-tool-releases v114
Measured on the day
- 184 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 v152
- 451 of 739 remote MCP endpoints probed answered a handshake (61.0%), on 461 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 v36
How this is made
- 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.
Earlier days
2026-11-082026-11-072026-11-062026-11-052026-11-042026-11-032026-11-022026-11-012026-10-312026-10-302026-10-292026-10-282026-10-272026-10-262026-10-252026-10-24