# Orca adds one-click resume for Codex and Claude Code sessions > Source: > Published: 2026-07-24 15:44:09+00:00 On July 23rd, 2026, [Orca ADE (@orca_build)](https://x.com/orca_build) announced a new feature that lets users resume a coding session in a different AI agent harness with a single click. The update, posted in a four‑tweet thread on X, addresses a common friction point for developers who hit usage limits while using OpenAI's Codex or Anthropic's Claude Code for code generation. The first tweet in the thread reads, "Hit your usage limit in Codex/Claude Code mid‑task? Orca now lets you resume the exact same session in a different agent harness — one click. Codex ⇄ Claude Code, full context intact🔥" and includes a short video demonstration. The video, hosted on X, shows the user interface where a "Resume" button appears once the platform detects a quota breach, allowing the same prompt history to be transferred to an alternative model without losing context. The feature was tracked in the project's public issue tracker. A user identified as @wngli85258651 (Rancho) opened issue #10250 on the Orca GitHub repository to request better session naming for easier locating. The issue, filed at [github.com/stablyai/orca/issues/10250](https://github.com/stablyai/orca/issues/10250), now serves as the official backlog item for the one‑click resume capability. Orca's maintainers marked the issue as filed and indicated that work is in progress. While the announcement does not include quantitative performance data, the functional change is clear: when an OpenAI Codex or Anthropic Claude Code job exhausts its allocated token or compute quota, Orca preserves the entire prompt‑completion chain and offers a seamless handoff to the other model. This preserves full context—variables, imports, and partially generated code—so the developer does not need to manually copy or recreate the session. The move reflects broader industry trends where multi‑model orchestration is becoming a standard feature of AI developer tools. Both OpenAI and Anthropic impose per‑request limits on their code‑focused models to manage compute costs and prevent abuse. Hitting those limits forces users to restart the workflow, often losing several minutes of work and breaking the mental model of the code generation process. By abstracting the limit handling, Orca lowers the operational overhead for developers and improves overall productivity. Orca itself is an open‑source framework that enables developers to build AI‑powered agents that can call each other, share state, and execute tasks autonomously. The repository lives under the Stably AI organization on GitHub, which has contributed multiple agent‑orchestration utilities to the community. The new resume capability expands Orca's utility from a pure orchestration layer to a more user‑friendly interface for interactive coding assistants. The release also hints at a broader strategy to position Orca as a bridge between competing code models. By supporting both Codex and Claude Code out of the box, Orca gives developers flexibility to switch between providers without rewriting integration code. This could be especially valuable for teams that benchmark model performance or cost on a per‑project basis. Early reactions on X were modest but positive. The original thread accumulated 249 likes, 14 retweets, and over 15,000 video views within hours of posting. Commenters such as @jamesperkins and @GoodFarmingAdam acknowledged the update, with the latter noting that an "experimental agent‑centric view is in‑progress," suggesting further UI enhancements may be on the roadmap. No financial details, funding rounds, or corporate announcements accompanied the feature release, and Orca's team remains un‑identified beyond the @orca_build handle. The lack of a formal press release underscores the project's community‑driven nature; updates are communicated directly to users on X and through the GitHub issue tracker. For developers already using Orca's agent platform, the one‑click resume feature can be accessed immediately by updating to the latest version of the Orca CLI or library, as indicated in the video demo. Detailed usage instructions are expected to be added to the project's documentation in the coming days. Overall, the addition of cross‑model session resumption marks a modest but practical step toward smoother AI‑assisted development workflows. By handling usage‑limit interruptions automatically, Orca reduces friction for teams that rely on large language models for code generation, allowing them to stay in the flow of writing and debugging rather than managing token budgets. **Sources**