Codesesh: A Local Browser for Your AI Coding Sessions

Codesesh, a new open-source tool, gives developers a unified local browser to view and search their AI coding agent sessions across multiple platforms — including Claude Code, Cursor, Codex, and OpenCode. The project aims to solve a growing pain point: as developers juggle multiple AI coding assistants, tracking past conversations becomes increasingly difficult.

Creator Xing Kaixin built the tool after struggling to share and revisit his own AI coding workflows. 'I originally wanted to share my process of working with coding agents with friends,' Xing explained, noting that reviewing specific past sessions was frustrating without a dedicated viewer.

What Codesesh Actually Does

Codesesh scans locally stored AI coding sessions and presents them in a unified, aggregated, and visual interface. Instead of digging through individual tool directories or fragmented log files, developers get a single dashboard for all their agent conversations.

The tool currently supports 5 platforms:

• Claude Code — Anthropic's CLI-based coding agent
• Cursor — the AI-first code editor
• Codex — OpenAI's coding agent
• OpenCode — the open-source terminal coding assistant
• Kimi — Moonshot AI's assistant (popular in Asia)

Codesesh evolved from 2 earlier utilities the creator had built — agent-dump and agent-view — which handled exporting and viewing sessions separately. Codesesh merges both functions into a single, more polished experience.

Why Session History Matters for AI-Assisted Development

Developers increasingly rely on AI coding agents for complex, multi-turn problem solving. A single session might span dozens of exchanges covering architecture decisions, debugging steps, and code generation. Losing track of these conversations means losing valuable context.

The problem compounds when teams use multiple tools. A developer might prototype with Cursor, debug with Claude Code, and experiment with Codex — each storing sessions in different formats and locations. Codesesh bridges that gap by normalizing session data across platforms.

How to Get Started

The project is available on GitHub and includes a feature overview at the project's documentation site. Since it runs entirely locally, session data never leaves the developer's machine — an important consideration for teams working with proprietary codebases.

Codesesh is still in its early stages, and the creator is actively seeking feedback from developers with similar needs. The tool is open source, making it straightforward to contribute support for additional AI coding agents as the ecosystem continues to expand.

A Growing Need in the AI Coding Tool Ecosystem

The launch of Codesesh reflects a broader trend: as AI coding assistants multiply, developers need better meta-tooling to manage their interactions with these systems. Tools like Claude Code, Cursor, and Codex have each attracted significant user bases in 2025, but interoperability and session management remain largely unsolved problems.

For developers who frequently switch between AI coding agents or want to audit their AI-assisted workflows, Codesesh offers a lightweight, privacy-friendly solution worth watching.