Daily digest · 2026-11-10 (UTC)

What broke for AI agents on 10 Nov 2026

11 incidents on 5 status pages.

Final. Built 2026-11-11 07:48 UTC, 7 hours after the day ended; this page will not change.

What broke

GitHub: “Incident with Git Operations and Pull Requests”critical impact · opened 16:10 UTC · resolved 17:55 UTC (104 minutes) · Actions | Pull Requestsstatus page ↗provider-incidents v109
GitHub: “Degraded availability for GPT-5.3 Codex and GPT-5.6 models in Copilot”major impact · opened Nov 9 23:16 UTC · resolved 00:05 UTC (49 minutes) · Copilot | Copilot AI Model Providersstatus page ↗provider-incidents v106
Cloudflare: “Changes to Cloudflare Workers are not producing Event Subscription events”minor impact · opened 01:53 UTC · identified as of Nov 11 00:00 UTC · Queues | Workersstatus page ↗provider-incidents v111
Cloudflare: “Account Analytics Filter Issue”minor impact · opened 08:02 UTC · resolved 08:45 UTC (43 minutes) · Analyticsstatus page ↗provider-incidents v108
Cloudflare: “Changes made to Workers may be delayed or disrupted”minor impact · opened Nov 9 16:15 UTC · resolved 01:46 UTC (9 h 31 min) · Workersstatus page ↗provider-incidents v107
Cloudflare: “Network Performance Issues in St. Louis, USA”minor impact · opened 22:50 UTC · resolved Nov 11 01:10 UTC (2 h 20 min) · St. Louis, United States - (STL)status page ↗provider-incidents v111
Cloudflare: “Prediction issue on OpenAI models via Replicate”minor impact · opened 18:35 UTC · resolved 19:06 UTC (31 minutes) · Replicatestatus page ↗provider-incidents v110
Vercel: “Elevated Sandbox and Blob request latency and errors”minor impact · opened 21:04 UTC · resolved 21:04 UTC (0 minutes) · Blob | Sandboxstatus page ↗provider-incidents v110
Anthropic: “Elevated errors on Fable 52 and Mythos 52”no impact level set · opened 21:47 UTC · resolved 21:47 UTC (0 minutes)status page ↗provider-incidents v110
OpenAI: “Elevated error rates with some U.S. business Mastercards”the feed gives no impact level · first seen by our reader Nov 7 00:19 UTC (the feed gives no start time) · resolved 16:06 UTCstatus page ↗provider-incidents v109
OpenAI: “Incorrect cost data in Costs API exports”the feed gives no impact level · first seen by our reader 06:20 UTC (the feed gives no start time) · resolved 07:02 UTCstatus page ↗provider-incidents v108

Shipped

No new major version of the SDKs and tools we follow.

33 other releases on the day

@anthropic-ai/claude-agent-sdk 0.3.322 (npm)released 01:57 UTC · previous 0.3.321agent-sdk-releases v125
@anthropic-ai/sdk 1.8.0 (npm)released 16:38 UTC · previous 1.7.0agent-sdk-releases v127
@google/genai 3.1.0 (npm)released 20:47 UTC · previous 3.0.0agent-sdk-releases v128
@langchain/core 1.3.7 (npm)released 18:44 UTC · previous 1.3.6agent-sdk-releases v128
ai 7.0.168 (npm)released 17:49 UTC · previous 7.0.166agent-sdk-releases v127
ai 7.0.169 (npm)released 20:18 UTC · previous 7.0.168agent-sdk-releases v128
anthropic 2.8.0 (pypi)released 16:33 UTC · previous 2.7.0agent-sdk-releases v127
anthropics/anthropic-sdk-python v2.8.0published 16:32 UTC · previous v2.7.0release notes ↗agent-tool-releases v125
anthropics/anthropic-sdk-typescript bedrock-sdk-v1.1.2published 16:33 UTC · previous bedrock-sdk-v1.1.1release notes ↗agent-tool-releases v125
anthropics/anthropic-sdk-typescript google-cloud-sdk-v1.1.4published 16:33 UTC · previous google-cloud-sdk-v1.1.3release notes ↗agent-tool-releases v125
anthropics/anthropic-sdk-typescript sdk-v1.8.0published 16:33 UTC · previous sdk-v1.7.0release notes ↗agent-tool-releases v125
anthropics/anthropic-sdk-typescript vertex-sdk-v1.1.3published 16:33 UTC · previous vertex-sdk-v1.1.2release notes ↗agent-tool-releases v125
anthropics/claude-code v2.1.322published 02:29 UTC · previous v2.1.321release notes ↗agent-tool-releases v123
BerriAI/litellm v1.103.11published 20:24 UTC · previous v1.108.4release notes ↗agent-tool-releases v126
BerriAI/litellm v1.104.9published 20:29 UTC · previous v1.103.11release notes ↗agent-tool-releases v126
BerriAI/litellm v1.108.5published 22:51 UTC · previous v1.104.9release notes ↗agent-tool-releases v126
google-gemini/gemini-cli v0.68.0published 21:17 UTC · previous v0.67.0release notes ↗agent-tool-releases v126
google-genai 3.1.0 (pypi)released 20:30 UTC · previous 3.0.0agent-sdk-releases v128
langchain-ai/langchain langchain-core==1.7.7published 04:08 UTC · previous langchain-core==1.7.6release notes ↗agent-tool-releases v124
langchain-ai/langchain langchain==1.4.10published 18:37 UTC · previous langchain==1.4.9release notes ↗agent-tool-releases v125
langchain-core 1.7.7 (pypi)released 04:08 UTC · previous 1.7.6agent-sdk-releases v125
litellm 1.108.5 (pypi)released 22:47 UTC · previous 1.108.4agent-sdk-releases v128
openai 3.49.0 (pypi)released 06:47 UTC · previous 3.48.0agent-sdk-releases v126
openai 7.55.0 (npm)released 06:58 UTC · previous 7.54.0agent-sdk-releases v126
openai/openai-node v7.55.0published 06:47 UTC · previous v7.54.0release notes ↗agent-tool-releases v124
openai/openai-python v3.49.0published 06:47 UTC · previous v3.48.0release notes ↗agent-tool-releases v124
vercel/ai @ai-sdk/[email protected]published 17:48 UTC · previous @ai-sdk/[email protected]release notes ↗agent-tool-releases v125
vercel/ai @ai-sdk/[email protected]published 17:53 UTC · previous @ai-sdk/[email protected]release notes ↗agent-tool-releases v125
vercel/ai [email protected]published 17:39 UTC · previous [email protected]release notes ↗agent-tool-releases v125
vercel/ai [email protected]published 17:45 UTC · previous [email protected]release notes ↗agent-tool-releases v125
vercel/ai [email protected]published 20:12 UTC · previous [email protected]release notes ↗agent-tool-releases v126
vercel/ai [email protected]published 17:48 UTC · previous [email protected]release notes ↗agent-tool-releases v125
vercel/ai [email protected]published 20:17 UTC · previous [email protected]release notes ↗agent-tool-releases v126

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 v168
468 of 800 remote MCP endpoints probed answered a handshake (58.5%), on 477 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 v40

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

2026-11-12 2026-11-11 2026-11-10 2026-11-09 2026-11-08 2026-11-07 2026-11-06 2026-11-05 2026-11-04 2026-11-03 2026-11-02 2026-11-01 2026-10-31 2026-10-30 2026-10-29 2026-10-28 2026-10-27 2026-10-26 2026-10-25 2026-10-24