AI Agent News: Releases, Launches and Usage Jumps, Live · page 35
-
Paseo passes 20K GitHub stars
-
Update your IDE to restore agent activity in Copilot usage metrics
GitHub is rolling out a fix to restore agent activity metrics in Copilot usage tracking after a bug caused agent metrics to appear lower than actual usage.
-
Cursor: Investigating service degradation — Cloud Agents
Major impact; resolved after 16 min.
-
Building Git infrastructure for agent-scale development
GitHub describes infrastructure improvements to handle increased write volume from agent-scale development.
-
Gemini CLI 0.63.0 released
fix(cli): display retry progress indicator during connection recovery; Changelog for v0.61.0-preview.1; Changelog for v0.61.0
-
Introducing Mistral Large 4: Le chonk
Mistral released a preview of Mistral Large 4, a 1 trillion parameter model with 49 billion active parameters, available via API with open weights coming end of month.
-
Oh My Pi 18.7.0 released
Fixed interrupted runs so assistant message boundaries are emitted reliably, allowing subscribers to persist and recover the interrupted turn.; Added getOAuthCredentialProvider() to resolve login aliases, such as openai-codex-device, to the provider where…
-
Cursor: Investigating service degradation — Cloud Agents and Grok Bot
Major impact; resolved after 2 h 39 min.
-
Building a context-aware AI assistant on AgentCore and OpenClaw
Tutorial on building a context-aware AI assistant using OpenClaw on Amazon Bedrock AgentCore, with persistent memory across conversations.
-
Cursor: Investigating service degradation — Security Reviewer Agents
Major impact; resolved after 1 h 7 min.
-
Paperclip 2026.1005.0 released
Operator UI snippet injection is removed. The PAPERCLIPCLOUDUISNIPPET and PAPERCLIPCLOUDUISNIPPETB64 settings no longer inject HTML into static or development pages. Operators who relied on them…; Keyboard shortcuts are always on, and their settings are…
-
TesterArmy e2e 0.18.0 released
#851 dc2afca Thanks @KrzysztofMoch! - e2e mcp sessions are now headless by default, the same as e2e run and e2e explore, and the agent decides per session: opensession {headed: true} shows the…; #790 0d1ed15 Thanks @okwasniewski! - Breaking: e2e and every…
-
Langfuse 4.53.0 released
feat(ai-gateway): expose resolution context in response headers, spans and logs; fix(web): zinc neutral tokens, surface tokens, observation line and fill colours; fix(evals): skip Bedrock URL media interpolation (+1 more release that day)
-
Claude Agent SDK: Elevated errors for Claude Opus 5.5
Major impact; resolved after 18 min.
-
Claude Code: Elevated errors for Claude Opus 5.5
Major impact; resolved after 18 min.
-
Claude Cowork: Elevated errors for Claude Opus 5.5
Major impact; resolved after 18 min.
-
Claude Design: Elevated errors for Claude Opus 5.5
Major impact; resolved after 18 min.
-
Claude in Chrome: Elevated errors for Claude Opus 5.5
Major impact; resolved after 18 min.
-
Advancing computer use with Ironclad
OpenAI and Ironclad are training and evaluating AI agents on complex contracting workflows to advance computer use capabilities.
-
Junie 3579.5 released
UX changes: trusted project on start, Notification Center, in-CLI notifications, removed the ○ ANSWER heading, mid-run steering; Fixed: no more hang on a locked system keyring; /review starts right away when the prompt names a commit, branch, range or PR…
-
Ruflo 3.53.0 released
Intelligence views: a memory map (a hash layout, labelled as such, not a similarity map), a learning pipeline and router picture, explain-a-recall with a pattern lifecycle, and memory health.; Workflows board: phases, agents, model, tokens and time per run…
-
Codex Security 0.2.0 released
Threat models are now saved with Standard, Deep, and Diff scan results and generated security policies. Export a saved model without another model call using codex-security export --artifact…; Scans now check for exposed credentials in source code, including…
-
OpenHands 1.25.0 released
feat: outline the back navigation control on settings sub-pages; feat(settings): bulk-add provider models as LLM profiles; feat: add direct-prompt Model Router settings
-
Mistral Large 4 now available on AI Gateway
Mistral Large 4, an open-weight multimodal model, is now available on Vercel's AI Gateway and accessible via the AI SDK.
-
Remote control for local agents
Cursor iOS app now allows users to see and control local agents running on their computer remotely.
-
DeepSeek Harness downloads down 42% in a week
316K npm and PyPI downloads in the 7 days to Oct 6, against 540K the week before.
-
OpenShell downloads up 114% in a week
87.3K npm and PyPI downloads in the 7 days to Oct 6, against 40.7K the week before.
-
Deep Agents passes 30K GitHub stars
-
OpenHands passes 90K GitHub stars
-
OpenShell gained 4,275 GitHub stars in a week
Now at 15.4K stars, against about 613 a week over the month before.
-
Strands Agents 1.58.0 released
feat(ts): port mcprouter vended tool to TypeScript (#4448) (ec86d614d); feat(bidi): normalize token usage events (#4886) (3973d1cb8); feat(multiagent): migrate agent axis policies to SDKs (#4714) (d87a0718a)
-
Secret scanning adds detectors for Lovable, Supabase, and more
GitHub secret scanning now detects secrets from Lovable Labs, Pydantic Services, and Supabase.
-
Opus 5.5 agents discover two room-temperature magnetic semiconductor candidates
Opus 5.5 agents discovered two room-temperature magnetic semiconductor candidates.
-
v0: Some users unable to access v0
Major impact; resolved after 1 h 13 min.
-
Zapier Agents: Logging into Zapier using Facebook or Microsoft SSO shown a blank page after authentication.
Major impact; resolved after 51 min.
-
Bland AI: Variable Extraction Model Routing Failures
Major impact; resolved after 0 min.
-
OpenAI "rogue" agent activities found on Wikimedia projects
OpenAI agent activities were found on Wikimedia projects.
-
Junie 3651.1 released
(+10 more releases that day)
-
Supercharge regulated workloads with Claude Code and Amazon Bedrock
Claude Opus 5.5 and Sonnet 5.5 are now available on Amazon Bedrock in GovCloud regions for use with Claude Code on regulated workloads.
-
New agent skill: Amazon SageMaker optimized generative AI inference for your coding agent
Amazon SageMaker introduces an aws-ai-ml skill for coding agents like Kiro and Claude Code to optimize generative AI inference.
This list shows notable items; the sections above show everything of their kind, including patch releases of smaller projects and low-impact incidents. Summaries of outside articles are written automatically from the source's own text and may miss nuance, so follow the link before relying on one.