Releases·Tool Release
Codesm: Claude Code, Codex, and your own models in one terminal
A developer building under the name Codesm released a terminal workspace that runs Claude Code, Codex, and API or local models side by side with a custom memory layer. The tool saves project history locally and passes relevant context between agents when users switch, addressing scattered conversation history across coding agents.
2 engagement1 sourceMon, Sep 7, 2026, 02:06 PM
AI Overview
- Codesm puts Claude Code, Codex, and API/local models in one terminal with a memory layer that passes project context when switching agents
- Codesm saves project history locally, is written in Python, MIT licensed, and installable from source on GitHub
- Its native agent used GPT-6 Astra to build Lunar Courier, a browser game with scoring, shields, and pause/resume; the demo video runs build excerpts at 4x speed and verification occurred after the agent hit its token budget
Related
Products·launched
VibeCode: multi-agent coding CLI wrapper for Windows
VibeCode is a native Windows desktop app that wraps multiple agentic coding CLIs (Claude Code, OpenAI Codex, Kimi Code, Grok) in a single UI. It launches each CLI as a child process, preserving existing logins and MCP servers, and allows running multiple agents on the same codebase simultaneously. It solves the problem of managing multiple coding agents across different tools for developers who want a unified interface.
VibeCode is a native Windows desktop app that wraps multiple agentic coding CLIs (Claude Code, OpenAI Codex, Kimi Code, Grok) in a single UI. It launches each CLI as a child process, preserving existing logins and MCP servers, and allows running multiple agents on the same codebase simultaneously. It solves the problem of managing multiple coding agents across different tools for developers who want a unified interface.
5 engagement2 sourcesredditSat, Jul 25, 2026, 12:25 AM
Products·building
AgentsRoom: Control room for multiple AI coding agents with auto model selection
AgentsRoom is a control room for managing multiple AI coding agents simultaneously, addressing the chaos of juggling many terminal tabs. It provides a unified interface to monitor agent status, switch between tasks, and automatically selects the best model for each task, improving efficiency for developers using tools like Claude Code and Cursor.
AgentsRoom is a control room for managing multiple AI coding agents simultaneously, addressing the chaos of juggling many terminal tabs. It provides a unified interface to monitor agent status, switch between tasks, and automatically selects the best model for each task, improving efficiency for developers using tools like Claude Code and Cursor.
3 engagement1 sourceredditFri, Aug 21, 2026, 04:43 PM
Releases·Tool Release
Developer builds Codex plugin to orchestrate Claude Code with Opus, Haiku, Sonnet
A developer created an open-source Codex plugin called codex-claude-orchestrator that lets Codex orchestrate a persistent Claude Code session, routing work to different Claude models: Opus maintains context, while Haiku and Sonnet handle bounded tasks. The tool aims to optimize model usage by assigning the strongest model only to work that truly requires it.
A developer created an open-source Codex plugin called codex-claude-orchestrator that lets Codex orchestrate a persistent Claude Code session, routing work to different Claude models: Opus maintains context, while Haiku and Sonnet handle bounded tasks. The tool aims to optimize model usage by assigning the strongest model only to work that truly requires it.
1 engagement1 sourceredditTue, Jul 21, 2026, 01:14 PM
Releases·Tool Release
Memcode launches terminal coding agent with persistent memory
Memcode, a developer platform in public beta, launched a terminal coding agent inspired by Claude Code that works across frontier and open-weight models. It features persistent memory and state recall to behave like a human coworker, plus chat, reusable agents, DataHub, and a website generator.
Memcode, a developer platform in public beta, launched a terminal coding agent inspired by Claude Code that works across frontier and open-weight models. It features persistent memory and state recall to behave like a human coworker, plus chat, reusable agents, DataHub, and a website generator.
3 engagement1 sourcehackernewsWed, Aug 5, 2026, 02:44 AM
Products·building
Multi-provider coding agent harness with per-session model routing
A harness that routes coding agent sessions across Claude Code, Codex, and Ollama, allowing users to pick a model per session. It abstracts sessions and history, supporting multiple execution engines and local models, solving the problem of model obsolescence by avoiding vendor lock-in.
A harness that routes coding agent sessions across Claude Code, Codex, and Ollama, allowing users to pick a model per session. It abstracts sessions and history, supporting multiple execution engines and local models, solving the problem of model obsolescence by avoiding vendor lock-in.
8 engagement1 sourceredditFri, Aug 7, 2026, 05:42 PM