The latest AI news and updates of Warp — Agentic development environment combining terminal, code editor, and AI agents for software development. Covering Warp's latest product updates from the past 90 days.
WarpJun 10Warp has integrated Anthropic's Claude Fable 5 model into its agentic development environment. This provides developers with a model capable of Mythos-level performance for autonomous, goal-oriented tasks, enhancing multi-step agent workflows.
WarpJun 5Warp's AI Agent now supports message queueing, allowing users to pre-load instructions for continuous execution. This update enables developers to queue, re-order, and auto-queue prompts, ensuring uninterrupted multi-step tasks without waiting for the agent to complete each action.
WarpJun 4Warp is introducing Agent Memory, a shared context system designed to help AI agents retain information across different sessions, tools, and teams. This system enables agents to learn from past interactions and avoid repeating errors, enhancing their efficiency and effectiveness in collaborative development environments.
WarpJun 2Warp has released an SSH extension that enables its visual file tree, codebase indexing, and autonomous coding features on remote macOS and Linux hosts. By replacing legacy shell commands with a native diff tool, the update allows AI agents to reason about and modify remote codebases with local-level precision.
WarpMay 29Warp introduced native commit, push, and pull request buttons to its Code Review panel, allowing developers to ship agent-generated code without leaving the terminal. This update completes the transition from AI-assisted prompting to production delivery within a single agentic environment.
WarpMay 28Warp integrated Anthropic's Claude Opus 4.8 and 4.8 Fast into its agentic development environment. The update shifts the focus from single-turn code generation to longer agent runs where models plan, execute, and review their own work.
WarpMay 27Warp introduced a setting that automatically transitions active terminal agent conversations to its Oz cloud platform when a user closes their laptop. This shift moves AI coding from ephemeral local sessions to persistent background services, ensuring long-running tasks like refactors aren't interrupted by hardware sleep.
WarpMay 26Warp added native support for OpenRouter, allowing developers to connect their own API keys to access a wide range of frontier and open-weight models directly from the terminal. This integration enables users to bypass platform credits by routing agent requests through their personal OpenRouter accounts for better pricing and uptime.
WarpMay 22Warp launched support for custom inference endpoints and Bring Your Own API Key (BYOK) for OpenAI, Anthropic, and Google models. This allows developers to bypass internal AI credits and use their own provider accounts for agentic terminal workflows while keeping keys stored locally.
WarpMay 21Warp developers can now move active terminal agent sessions to the cloud by prefixing prompts with the & symbol. This shift allows long-running autonomous tasks to continue executing independently of the local machine while providing a link for remote monitoring.
WarpMay 20Warp integrated Google's Gemini 3.5 Flash into its terminal-based AI agent to optimize autonomous development tasks. The update provides a high-speed, cost-effective alternative for multi-step coding loops where low latency is more critical than raw reasoning depth.
WarpMay 19Warp introduced multi-agent orchestration to its Oz platform, allowing developers to delegate tasks across Claude Code, Codex, and the Warp Agent simultaneously. A new handoff feature lets users move active terminal agent sessions to the cloud to run persistently after closing their local environment.
WarpMay 6Warp moved its terminal configuration to a settings.toml file that is now fully accessible to autonomous AI agents. By exposing settings as structured data, developers can use coding agents to customize their environment through natural language rather than manual menus.
WarpMay 6Warp released its internal library of Agent Skills to provide open-source instruction sets for Terraform, dbt, and documentation tasks. By sharing the procedural knowledge used by its own engineering team, Warp is standardizing how autonomous agents navigate complex development environments.
WarpMay 5Warp introduced an auto (open-weights) setting that routes terminal tasks to frontier-level open models based on request complexity. This update allows developers to maintain high performance during autonomous workflows while significantly reducing token expenses compared to the platform's high-end genius mode.
WarpMay 2Warp introduced a /feedback command that uses an AI agent to draft and post high-quality GitHub issues directly from the terminal. This skill bridges the gap between identifying bugs and fixing them by guiding users through a report-and-repair loop within the same agentic environment.
WarpApr 28Warp has open-sourced its client codebase under a dual MIT/AGPL license and repositioned itself as a platform for orchestrating AI agents. This shift moves the terminal from a standalone tool to a unified harness for managing autonomous coding agents like Claude Code and Codex.
WarpApr 25Warp added OpenAI's GPT-5.5 to its terminal-based AI agent, delivering a 40% speed increase over the previous model version. The update focuses on improving the responsiveness and cost-effectiveness of autonomous development loops like site migrations and code verification.
WarpApr 13Warp now sends desktop and in-app notifications when its AI agent is blocked by manual input requirements like password prompts. This update prevents autonomous agents from stalling silently, allowing developers to step away from long-running tasks without needing to manually monitor terminal progress.
WarpApr 6Warp's AI agent now uses a dedicated tool to present users with multi-choice questions during the planning and feedback stages of a task. This shift from open-ended chat to structured UI options keeps developers in the loop without requiring manual text input for every decision.
WarpApr 4Warp now supports native code reviews and inline commenting for AI agents like Claude Code and Codex directly in the terminal. This creates a structured human-in-the-loop workflow that allows developers to provide granular feedback to agents without leaving their development environment.
WarpMar 31Warp introduced a new @ conversation command that allows developers to attach the history of past AI interactions as context for their current agent session. This enables users to reset an agent's focus while selectively retaining critical logic or debugging steps from previous threads.
Frequently asked questions