AI Coding Live News

The OpenAI Developers account shared a post by Steve (@stevendcoffey), who works on the API at OpenAI, listing five Agents API updates this week: Codex-style Auto Review of commands before execution, compaction that runs in the background, an experimental time-awareness feature, updating a session's tools at any time, and automatic caching of system prompts across all sessions. As of writing, the official API changelog does not list these items yet.

Anthropic published Claude Code v2.1.314 on GitHub at 01:06 UTC on Oct 30. The official CHANGELOG lists 119 entries, more than half of them fixes. The new features lean enterprise: permissions.blockWritesOutsideWorkingDirectories makes Write, Edit and NotebookEdit refuse paths outside the working directories in every permission mode; Claude apps gateway upstreams can be limited to IdP groups, and Bedrock upstreams can log which guardrail fired; idle compaction now also runs in VS Code and the Code tab of Claude Desktop. It also fixes sessions failing repeatedly when an admin caps effort, and claude -p exiting 0 with no output when startup failed.

OpenAI's Codex lead Tibo says GPT-6.1 Sol's default context in Codex and ChatGPT Work goes from 272K to 500K (~80% more) to cut compactions on long tasks, with no long-context fees when usage comes from a ChatGPT subscription. Separately, the official Codex changelog says GPT-5.5 retires from ChatGPT and Codex on all plans on October 14 (the API keeps it).

As autonomous agents execute complex multi-step workflows across terminal consoles and web browsers, context window expansion and escalating token costs impose severe operational bottlenecks. Conventional compaction methods rely on text summarization or token pruning, both of which discard crucial state variables and induce semantic drift over extended sessions. Researchers from HKUST and Alibaba present REMORY (arXiv:2610.11287), introducing a residual memory paradigm designed for sequence-level context compaction. REMORY combines lossy natural language summaries with bounded sets of learned soft memory tokens, effectively forming a residual connection in the sequence dimension. While text summaries convey high-level narrative continuity, the soft memory tokens reconstruct lost fine-grained execution artifacts, API arguments, and intermediate environment states. On challenging benchmarks including BrowseComp and Terminal-Bench 2.1, REMORY retains 97% of full-context task performance while occupying only 5.2% of the original input token footprint, slashing redundant tool retries and loops by more than 40%.

@CompleteSkeptic (Diogo Almeida, TypeSafe; RLHF background) published public design notes on TypeSafe × coding agents (Google Doc), covering routing, context compaction, subagents, skills, and MCP tradeoffs, and invited the community to experiment. The post drew ~930 likes / ~56k views the same day.

@aiRobertDaily highlighted fast-jev-compaction, a Claude Code plugin that replaces traditional LLM summarization with Jev-driven, decision-based context compaction. It quickly filters tool calls and results, drops or truncates stale content, and keeps only the necessary parts of valuable text; a related Codex-facing Jev context server was also noted. The aim is lower token cost and faster delivery.

LM Studio (@lmstudio) introduced Introspection and Session References for its Bionic agent: the agent can search its own session history to stay useful after many compactions, and the composer lets you @-mention other sessions to continue work across chats.

Grok Build shipped v1.0.33 with stronger MCP, memory, and long-session reliability: MCP tool results can include structured JSON, cancel/timeout notifies the server to stop work, /memory supports two-press delete for real notes, and compaction checkpoints survive the 30-day cleanup so long agent workflows keep context. Elon Musk amplified the update.

Sourcegraph introduced a semantic context compaction engine for Cody. Designed for deep call-stack comprehension, the algorithm prunes boilerplate while strictly retaining function signatures, type contracts, and critical execution paths. This compresses token usage by 60% and increases multi-package code synthesis accuracy by 34%.