SShortSingh.
Back to feed

Controlling Claude Code and Codex from a phone: official options and a self-hosted app

0
·1 views

Long-running coding agents often pause for confirmation and sit idle until the user returns to their desk. Anthropic's Claude Code offers a Remote Control mode that lets users continue a session from claude.ai/code or the mobile app, requiring the same account and a subscription plan. OpenAI's Codex can be paired with the ChatGPT mobile app via the desktop app, keeping the computer awake and online. The author, who runs several agents on personal machines, built TaroBird, an open-source, self-hosted chat app that wraps official CLIs locally and lets bots message each other. The app also offers remote access through Tailscale or an encrypted relay, though its relay does not run bots or store chats.

Read the full story at DEV Community

This is an AI-generated summary. ShortSingh links to the original source for the complete article.

Discussion (0)

Log in to join the discussion and vote.

Log in

Related stories

0
ProgrammingDEV Community ·

Eclipse Enclave 1.0 Launches Open-Source Sandbox for AI Coding Agents

The Eclipse Foundation released Enclave 1.0 on 19 November 2026, a free, MIT-licensed sandbox that runs each AI coding agent in its own isolated container. Users control which files, network hosts and secrets the agent can access, letting it work autonomously without prompts while keeping the host machine protected. The tool is vendor-neutral and ships with six agents built in, including Claude Code, Codex and GitHub Copilot CLI. It grew out of internal tooling at EclipseSource, which contributed it to the Foundation in July 2026. A gateway component filters outbound traffic against an allowlist and logs both allowed and blocked attempts for auditing.

0
ProgrammingDEV Community ·

Only 41 of 90 new article pairs are ready to publish, all for Zenn

A production ledger check at commit 6fe4649 on September 11, 2026 found 136 Japanese and English files forming 68 bilingual pairs, but only 41 pairs are fully complete with resolved URLs. The remaining 27 pairs lack canonical article IDs, and 22 candidates still need body verification. All 41 completed pairs target Zenn, while none are complete for Qiita. Under a hypothetical schedule of one Zenn pair per day, the stock covers 41 consecutive days, but allocating two Zenn pairs daily yields only 20 days. The existing publishing schedule contains no rows on or after September 11, so none of the new candidates has been scheduled yet.

0
ProgrammingDEV Community ·

Microsoft Adds Knowledge Graphs to Foundry IQ to Improve Multi-Hop Retrieval

Microsoft has added Knowledge Graphs and a Microsoft-Retriever-1 component to its Foundry IQ product, aiming to reduce reliance on large frontier models for planning multi-hop retrieval over flat vector chunks. Starburst separately launched Context Studio, which exposes governed semantic context to agents over MCP to limit conflicting business definitions. Both moves point to a growing view that vector search alone is insufficient and that prompting frontier models with everything is slow and costly. The author, who is building an open-source self-hosted alternative called PipesHub, reports practical lessons such as combining deterministic structural links with selective entity linking rather than relying solely on costly LLM-based triplet extraction. The author also describes implementing separate identity handling for service-account tokens and encountering OAuth Dynamic Client Registration failures with Dropbox's MCP server, arguing that pre-registered clients should be supported alongside existing methods.

0
ProgrammingDEV Community ·

NVIDIA's DLSS5 Aims to Deliver Richer Visuals Without Heavier Hardware Demands

The article, an unofficial side-story piece from DEV Community, examines DLSS5, an AI-driven upscaling technology that renders lower-resolution images in real time. It argues that DLSS5 lets games display detailed environments, vivid colors, and intricate textures while keeping frame rates smooth. The piece highlights a freestyle version that lets developers tailor graphics settings to their preferred balance of quality and performance. It also describes performance optimization, where algorithms dynamically adjust rendering to reduce stutter, even on lower-end hardware. The author presents these advances as a way to end the long-standing trade-off between visual fidelity and gameplay performance.