AICoder

AI Coding Live News

First-hand AI coding updates from 𝕏/Twitter frontier labs and researchers, distilled by AI with entity benchmarks.

Explore live on 𝕏 ↗

AI Coding Live News

T
Threat Intelligence (Symantec)@threatintel·
👀75
🔥82
View on 𝕏 ↗

Symantec finds a grey-market relay that installs the real Codex and Claude Code CLIs, then turns off sandboxing and routes every request through its own servers

On Oct 29 Symantec's Threat Hunter Team disclosed an unauthorized AI relay seen at about a dozen organizations since early 2026, mostly in mainland China. Users install the genuine Codex or Claude Code CLI via npm, then run a one-line script from the reseller that rewrites ~/.codex/config.toml or ~/.claude/settings.json, points base_url at relay domains, swaps in relay-issued keys, sets the sandbox to danger-full-access and disables telemetry. Symantec says this is not a CLI vulnerability and it has seen no abuse so far.

Symantec finds a grey-market relay that installs the real Codex and Claude Code CLIs, then turns off sandboxing and routes every request through its own servers

Key Takeaways

  • About a dozen affected organizations, mostly in mainland China, plus Chinese or home-market units of South Korean, Taiwanese and Malaysian firms in semiconductors, electronics, pharma, shipping and petrochemicals
  • A one-line iwr | iex or curl | bash command fetches a fileless script from lookalike domains such as vibe.aiok[.]me, installs the genuine CLI, then rewrites its config, so it looks like a normal setup step
  • The Codex sandbox is set to danger-full-access with network access on, and DISABLE_TELEMETRY=1 and DISABLE_ERROR_REPORTING=1 are set
  • Relay-issued Claude Code keys use the sk-ant-oat01- prefix of subscription OAuth tokens, which Symantec says is consistent with, but not proof of, pooled subscription resale
  • Symantec says this is not a vulnerability in Codex or Claude Code and it has no evidence the operator has abused its access; the risks are potential
Read full story
Linked Models & Agents:
🤖 OpenAI Codex🤖 Claude Code🌐 Symantec🌐 OpenAI🌐 Anthropic
#Security#AI relay#Codex#Claude Code#Sandbox#supply chain
A
Anthropicgithub.com·
👀65
🔥82

Claude Code 2.1.304 adds guided /auto-mode-setup and warns Pro/Max users before resuming an inactive conversation would eat into the 5-hour usage limit

Claude Code v2.1.304 (Oct 19 UTC) adds /auto-mode-setup, a guided setup that reads your project and recent sessions to teach auto mode about your environment, now also on Bedrock, Vertex, Foundry, Claude apps gateway and telemetry-disabled sessions. It adds a read-only /claude-api managed-agents-secure check against Anthropic's security guide, warns Pro/Max users when resuming an inactive conversation would use a noticeable share of the 5-hour limit, and fixes 40+ issues including claude --worktree deleting a worktree and branch without asking, which could lose commits.

Key Takeaways

  • New /auto-mode-setup reads your project and recent sessions to teach auto mode your environment, now on Bedrock, Vertex, Foundry, Claude apps gateway and telemetry-disabled sessions
  • New read-only /claude-api managed-agents-secure compares a Managed Agents agent's settings with Anthropic's published security guide
  • Pro/Max plans get a warning when resuming an inactive conversation would use a noticeable share of the 5-hour usage limit
  • Mods gain decider: "person" and e.isReadOnly in tool.check, so only a human permission dialog can approve certain calls
  • 40+ fixes, incl. claude --worktree removing the worktree and branch without asking (could lose commits) and CLAUDE_CODE_AUTO_COMPACT_WINDOW=200k silently becoming 100K
Read full story
Linked Models & Agents:
🤖 Claude Code🤖 Managed Agents🌐 Anthropic
#Claude Code#Anthropic#auto mode#Managed Agents#usage limits#Worktree
P
Peter McCrory@PeterMcCrory·
👀41
🔥86
View on 𝕏 ↗

Anthropic publishes internal employment report: junior share of technical hires fell from 27% to 9% as agent hours reached 6x employee hours by August

On Oct 16 Anthropic economists published "How employment is changing at Anthropic," drawing on internal HR, agent telemetry, commits, calendars and hiring records. The junior share of technical hires fell from 27% to 9%; by August employee-directed interactive agent hours were 6x employee hours; commits per engineer are 5x their pace six months ago; and Claude sessions spend about 80% of their time waiting on humans, making human attention the bottleneck.

Anthropic publishes internal employment report: junior share of technical hires fell from 27% to 9% as agent hours reached 6x employee hours by August

Key Takeaways

  • Junior share of technical hires fell from 27% to 9%; junior share of technical headcount fell from 21% to 14% since early 2025 (Anthropic data)
  • Total agent hours passed human hours in March 2026; by August employee-directed interactive agent hours alone were 6x employee hours
  • Commits per engineer are 5x their pace six months ago; comparing 192 adopters with 74 non-adopters, agent-written code showed no difference in later deletion or revert rates
  • Interactive Claude sessions spend about 80% of their time waiting for humans; meta-work rose from 10% to 22% of sessions
  • Staff expect AI to do 75% of their work by 2030 vs 28% for other Americans; Anthropic responds with an early-warning system and internal redeployment
Read full story
Linked Models & Agents:
🌐 Anthropic
#Anthropic#Coding Agent#Anthropic Institute
O
OpenAI Developers@OpenAIDevs·
👀80
🔥90
View on 𝕏 ↗

OpenAI: GPT-6 Astra and 6.1 Sol for software maintenance — 1,400+ cleanup PRs and 800k+ lines removed

An OpenAI developers blog shows how Core Experimentation uses Codex with GPT-6 Astra to clean dead code and stale Statsig experiments—evidence from Slack, telemetry and flags into small draft PRs. Official: 1,400+ maintenance PRs and 800k+ lines deleted over ~two part-time weeks. GPT-6.1 Sol Ultra reached ~92% of Astra Ultra’s accepted findings at ~14% API cost (vendor-reported).

OpenAI: GPT-6 Astra and 6.1 Sol for software maintenance — 1,400+ cleanup PRs and 800k+ lines removed

Key Takeaways

  • Official: 1,400+ maintenance PRs and 800k+ lines deleted over ~two part-time weeks
  • One image 3.31GB→1.72GB (~48%); ~400 visual snapshot variants removed
  • Same audit: Sol Ultra 131 findings/$15.12 vs Astra Ultra 142/$109.06 (vendor; ~92% findings, ~14% cost)
  • A Permanent all-features override removed after human review; ~0.42% CPU estimated
  • CI follow-ups (~30 PRs): median ~10→4.5 min; one missing mock cut 63 tests 592.6s→29.8s
Read full story
Linked Models & Agents:
⚡ GPT-6 Astra⚡ GPT-6.1 Sol⚡ GPT-5.6 Sol🤖 OpenAI Codex🌐 OpenAI
#OpenAI#Codex#GPT-6 Astra#GPT-6.1 Sol#Statsig#CI
ADSponsored
C
Claude Devs@ClaudeDevs·
👀92
🔥88
View on 𝕏 ↗

Claude Platform opens OpenTelemetry tracing beta: Messages API, Managed Agents and SDK spans flow to your own observability stack

On 2026-10-14 Anthropic put OpenTelemetry tracing for Claude Platform into public beta. Admins add a telemetry destination in the Claude Console and the API exports spans for Messages API requests and Managed Agents sessions server-side with no code changes, while the Claude SDKs export client-side spans for requests and tool-runner tools. Spans follow the OpenTelemetry gen_ai conventions and default to metadata (duration, status, token counts, model, tool names), with an optional full-content mode.

Claude Platform opens OpenTelemetry tracing beta: Messages API, Managed Agents and SDK spans flow to your own observability stack

Key Takeaways

  • Two paths: a server-side destination via Console/Admin API (Messages API + Managed Agents, no code changes) or client-side spans from the Claude SDK (overview)
  • Destinations need OTLP/HTTP with protobuf over HTTPS on port 443; the Admin API uses the anthropic-beta: telemetry-destinations-2026-08-11 header (API guide)
  • SDK tracing covers Python, TypeScript, Go and Java (anthropic-java 2.72.0+), and runs on Claude API, AWS, Bedrock, Google Cloud and Microsoft Foundry, all in beta (SDK guide)
  • Metadata-only or full-content modes; ZDR customers can still export full content for Messages API requests (held in memory until delivered); Managed Agents is not ZDR-eligible
  • Anthropic warns span names and SDK options may change during the beta, so pin SDK versions
Read full story
Linked Models & Agents:
🤖 claude-managed-agents🤖 claude-agent-sdk🌐 Anthropic🌐 datadog🌐 grafana🌐 honeycomb🌐 langfuse
#Anthropic#Claude Platform#OpenTelemetry#Managed Agents#Claude SDK
V
Vercelvercel·
👀68
🔥90

Vercel AI SDK: [email protected]–7.0.126 speech telemetry + tool-approval clear; @ai-sdk/openai ultrafast service tier

Vercel shipped [email protected] (speech/transcription telemetry) through [email protected] (clear tool approvals on addToolOutput) on 2026-09-30—22 commits since covered [email protected]. Same day @ai-sdk/[email protected] adds OpenAI serviceTier: 'ultrafast' (docs: access-controlled, gpt-5.6-sol), plus Azure MAI speech/transcribe and MCP OAuth conditional invalidation patches in the tree.

Vercel AI SDK: ai@7.0.124–7.0.126 speech telemetry + tool-approval clear; @ai-sdk/openai ultrafast service tier

Key Takeaways

Read full story
Linked Models & Agents:
⚡ GPT-5.6 Sol🌐 vercel🌐 OpenAI
#Vercel AI SDK#[email protected]#Speech#Telemetry#Ultrafast#Tool approval#OpenAI#Release
D
Docker@docker·
👀57
🔥90

Docker cagent 1.145.0: ACP request traces via W3C traceparent/tracestate, server spans on 13 handlers, lean TUI scrollback fix

Docker shipped cagent (docker-agent) v1.145.0 on 2026-09-28: ACP requests now propagate W3C traceparent/tracestate with server spans for each of the 13 implemented protocol handlers, plus new lint cops and a lean-TUI fix that preserves scrollback during partial streaming tool calls.

Docker cagent 1.145.0: ACP request traces via W3C traceparent/tracestate, server spans on 13 handlers, lean TUI scrollback fix

Key Takeaways

  • Shipped v1.145.0 — ACP W3C traceparent/tracestate + 13 handler spans
  • Lean TUI keeps scrollback during partial streaming tool calls
  • New FieldsSeqLookup / SlicesConcat lint cops
  • Docs: docker.github.io/docker-agent
Read full story
Linked Models & Agents:
🤖 cagent🤖 Docker-Agent🌐 Docker
#Cagent#Docker#ACP#OpenTelemetry#TUI#Tracing#Coding agent
L
LiveKit@livekit·
👀90
🔥91

LiveKit Agents 1.8.3: Gemini 3.8 Live NON_BLOCKING default, GPTLive Azure, AssemblyAI Universal, safer tool handoffs

LiveKit published Python [email protected] on GitHub (2026-09-26; PyPI livekit-agents 1.8.3): Gemini 3.8 Live defaults tool_behavior=NON_BLOCKING and sends FunctionTool schemas as parameters_json_schema on the text API; OpenAI gains GPTLiveModel.with_azure and moves the Realtime inference client into agents core; STT adds AssemblyAI universal-3-6-pro plus language_confidence, and configurable ElevenLabs realtime chunk duration; voice fixes prevent parallel tools from canceling handoffs, pair tool outputs by call_id, forward frames as-is in fallback/stream adapters, and fix LiveAvatar interruption/plugin registration. Same-day JS @livekit/[email protected] deepens agent_turn telemetry and adaptive interruption.

LiveKit Agents 1.8.3: Gemini 3.8 Live NON_BLOCKING default, GPTLive Azure, AssemblyAI Universal, safer tool handoffs

Key Takeaways

  • Shipped: [email protected] / PyPI 1.8.3; JS @livekit/[email protected] same day
  • Gemini 3.8 Live: default NON_BLOCKING tool_behavior; FunctionTool via parameters_json_schema on text API
  • OpenAI: GPTLiveModel.with_azure; Realtime inference client moved into agents core
  • STT: AssemblyAI universal-3-6-pro + language_confidence; configurable ElevenLabs realtime chunk duration
  • Voice correctness: parallel tools no longer cancel handoffs; pair tool outputs by call_id; adapters forward frames as-is
Read full story
Linked Models & Agents:
⚡ Gemini 3.8 Live⚡ GPT Live🤖 LiveKit Agents🌐 livekit🌐 Google DeepMind🌐 OpenAI🌐 assemblyai🌐 elevenlabs
#LiveKit Agents#voice AI#Gemini 3.8 Live#GPTLive Azure#AssemblyAI#Realtime#Python
C
CopilotKit@CopilotKit·
👀75
🔥88

CopilotKit 1.74.0: consume learned skills across SDK containers; Web Inspector targeted What's New + HUD telemetry

CopilotKit shipped v1.74.0 on 2026-09-25 (npm @copilotkit/react-core 1.74.0): SDKs can consume learned skills from multiple containers, expanding skill assets beyond a single runtime; Web Inspector gains targeted What's New notifications and launcher HUD impression/action tracking. Fixes: threads drawer no longer locks on unresolved entitlements, stop refreshing realtime credentials for sockets that never open, stop per-message state cloning / virtual-scroll tug-of-war in React, and a Memory policy that grants nothing no longer fails the run.

CopilotKit 1.74.0: consume learned skills across SDK containers; Web Inspector targeted What's New + HUD telemetry

Key Takeaways

  • Shipped: v1.74.0 / @copilotkit/[email protected]
  • Skills: consume learned skills from multiple containers across SDKs (#7384)
  • Inspector: targeted What's New (#6956) + launcher HUD impression/action tracking (#7376)
  • Stability: no realtime credential refresh for never-open sockets; empty-grant Memory policy no longer fails runs
  • React: stop per-message state cloning and virtual-scroll tug-of-war on long threads
Read full story
Linked Models & Agents:
🤖 CopilotKit🌐 copilotkit
#CopilotKit#learned skills#Web Inspector#React#agent UI#Memory policy
C
Cline@cline_dev·
👀85
🔥93

Cline Desktop v0.0.37 Released: Adds Built-in Updater & Highlights Dialog, Patches Bedrock Reasoning Config

Cline Desktop (@cline_dev) has shipped release v0.0.37. The update introduces a comprehensive lifecycle management experience, embedding a dedicated 'About' view within Settings that enables in-app update checks, one-click restarts, and inline changelog inspection. Users are greeted with an updated 'What's New' dialog highlighting recent architectural milestones including SSH remotes, Git worktrees, pull request telemetry, and parallel sub-agents. Crucially, the release patches an AWS Bedrock inference bug that triggered 'Unknown parameter: reasoningConfig' on OpenAI models, while adding a macOS diagnostic export shortcut.

Cline Desktop v0.0.37 Released: Adds Built-in Updater & Highlights Dialog, Patches Bedrock Reasoning Config

Key Takeaways

  • Introduces an in-app About settings pane featuring direct update checking, automated restart patching, and release notes.
  • Launches a unified What's New dialog summarizing SSH remotes, Git worktrees, PR composer status, and parallel sub-agents.
  • Fixes an AWS Bedrock regression causing OpenAI models to crash with 'Unknown parameter: reasoningConfig'.
  • Adds a direct macOS Help menu shortcut for one-click diagnostic log exporting.
  • Signed binaries for macOS, Windows, and Linux available directly on GitHub Releases.
Read full story
Linked Models & Agents:
⚡ GPT-6 Astra⚡ Claude Opus 5.5🤖 Cline (VS Code Agent)🤖 desktop-agent🌐 cline🌐 Amazon AWS (Nova)
#Cline#Desktop App#Coding Agents#Amazon Bedrock#Update#Open Source
ADSponsored
G
GitHub Changelog@GHchangelog·
👀58
🔥88
View on 𝕏 ↗

GitHub Copilot app adds OpenTelemetry: export agent traces to org monitoring via managed settings

GitHub Changelog (@GHchangelog) announced OpenTelemetry support in the Copilot app via enterprise-managed settings: admins export agent session activity to org-compatible monitoring to analyze sessions, investigate surprises, and manage telemetry centrally. Prompt/response content is excluded by default; configure telemetry in managed-settings.json with an export endpoint.

GitHub Copilot app adds OpenTelemetry: export agent traces to org monitoring via managed settings

Key Takeaways

  • Enterprise-managed settings enable Copilot app OpenTelemetry export.
  • Use cases: analyze agent sessions (model + tool calls), step-through unexpected behavior, centralize monitoring.
  • Prompt/response content excluded by default—review content-capture before enabling.
  • Set telemetry + endpoint in enterprise managed-settings.json. Concepts: OpenTelemetry for agent monitoring.
Read full story
Linked Models & Agents:
🤖 GitHub Copilot (Agent Mode / Codex)🌐 github🌐 Microsoft Azure AI
#GitHub Copilot#OpenTelemetry#Observability#Enterprise#agent monitoring#managed settings
d
d4rkninja@th3d4rkninja·
👀58
🔥85
View on 𝕏 ↗

Lattice Talk open-sources MCP so Cursor, Claude Code, and Codex can talk

@th3d4rkninja shipped Lattice Talk, an open-source MCP communication layer so Cursor, Claude Code, and Codex stop using the human as a message bus. Agents that share a session_id can discover each other, DM, join rooms, share project memory, coordinate across machines, and trace multi-agent workflows with OpenTelemetry. Aimed at MCP clients; open for PRs and API-credit sponsors.

Key Takeaways

  • Open-source MCP layer for Cursor / Claude Code / Codex interop.
  • Shared session_id enables discovery, DMs, rooms, project memory, cross-machine coordination.
  • Multi-agent workflows can be traced with OpenTelemetry.
Read full story
Linked Models & Agents:
🤖 Claude Code (Anthropic)🤖 Cursor Composer (Agent Mode)
#Lattice Talk#MCP#Multi agent#Claude Code#Codex#Cursor#open source#OpenTelemetry
G
Gentleman Programming@G_Programming·
👀40
🔥86
View on 𝕏 ↗

gentle-ai v2.9.0: RTK community tool, Claude Code telemetry, OpenCode reviewer relay

Gentleman Programming shipped gentle-ai v2.8.1–v2.9.0 and gentle-pi v2.6.1–v2.6.3. Highlights include Claude Code runtime telemetry that attributes models by public family patterns, OpenCode SDD/reviewer relay semantic decoding, and opt-in RTK as a community tool (attested install to ~/.local/bin/rtk, configures only selected agents, telemetry off by default; Windows unavailable). gentle-pi adds subagent_reconcile so a parent can reconcile uncertain acquire/settle remediations without launching an actor or exposing tokens.

Key Takeaways

  • gentle-ai v2.9.0 adds opt-in RTK community tool with attested install and selected-agent-only config.
  • Claude Code Stop usage telemetry is idempotent and attributes models by public family patterns.
  • OpenCode reviewer relay decodes Task wrappers semantically; gentle-pi adds subagent_reconcile for uncertain remediations.
Read full story
Linked Models & Agents:
🤖 Claude Code (Anthropic)🤖 opencode
#Gentle AI#Gentle pi#RTK#Claude Code#OpenCode#Telemetry#agent tooling
T
Telemetry@telemetrydev·
👀56
🔥84
View on 𝕏 ↗

Telemetry: instrument Cursor, Codex, Claude Code, pi, omp, and OpenCode agents

Telemetry (@telemetrydev) launched an open-source instrumentation stack (github.com/telemetry-dev/telemetry) for coding agents—Cursor, Codex, Claude Code, pi, omp, and OpenCode—using OpenTelemetry to track tokens, cost, latency, and errors.

Key Takeaways

  • Product: open-source Telemetry agent instrumentation (OpenTelemetry).
  • Coverage: Cursor, Codex, Claude Code, pi, omp, OpenCode.
  • Metrics: tokens, cost, latency, and errors.
Read full story
Linked Models & Agents:
🤖 Claude Code (Anthropic)🤖 openai-codex🤖 Cursor Composer (Agent Mode)
#Telemetry#OpenTelemetry#Cursor#Codex#Claude Code#agent observability
A
A Pale Blue Dot@yagyaansh·
👀48
🔥88
View on 𝕏 ↗

Turntrail: open-source VS Code extension for local deterministic handoff across Claude Code / Codex / Gemini CLI / Cursor

@yagyaansh launched free open-source VS Code extension Turntrail: local deterministic handoffs across Claude Code, Codex, Gemini CLI, and Cursor using real transcripts and workspace state — plus native session resume, multi-account switching, Codex usage/reset monitoring, and no telemetry.

Key Takeaways

  • Ship: free open-source VS Code extension Turntrail for multi-agent workflows.
  • Mechanism: local deterministic transcript/workspace handoff, no telemetry.
  • Extras: multi-account switching and Codex usage/reset monitoring.
Read full story
Linked Models & Agents:
🤖 Claude Code (Anthropic)🤖 openai-codex🤖 Cursor Composer (Agent Mode)
#Turntrail#VS Code#Claude Code#Codex#Gemini CLI#Cursor#Handoff#Open source