Celeborn Code is a tool designed to provide persistent memory and context management for AI coding agents working on software projects. It addresses the challenge of maintaining project context across sessions, ensuring that AI agents retain knowledge of what has been built, previous decisions, and upcoming tasks, thereby reducing the need for repeated explanations and avoiding lost progress. The tool operates primarily as a local command-line interface, requiring only Python and SQLite for installation, with no account necessary and all data remaining on the user's machine. Key features available in the local, offline mode include tiered file memory with automatic rehydration, per-turn capture, full-text search, a local kanban board for project management, token and dollar savings estimates, a multi-agent workflow supporting claim, touches, and outbox management, and compatibility with various AI coding agent harness adapters such as Claude Code, Grok, and Codex. Additional integrations include git-daemon sync and a GitHub App for drift checking and thread capture. The local board supervises itself to ensure continuous availability during sessions. For users requiring cross-device synchronization or team coordination, Celeborn Code offers hosted plans. The Pro plan enables real-time, hosted sync across devices, unlimited projects, a rich hosted board accessible via the web, and enhanced GitHub App integration. The Team plan builds on Pro by adding features for shared projects, organizational administration, shared context among team members, a multi-agent communication bus, and a fleet dashboard spanning multiple repositories. Enterprise options include SSO/SAML support and custom agreements. Celeborn Code is delivered as a CLI tool, installable via package managers such as Homebrew and Scoop, with a browser extension for Chrome available to integrate browser-based workflows. The tool is built in the open, with features and updates managed transparently through its own kanban board interface. Pricing includes a free local tier with no account required, a Pro plan at $8 per seat per month, and a Team plan at $12 per seat per month, with enterprise pricing available on request.
celeborn-code sits in PulseGate's Autonomous agents & workflows category. It focuses on maintaining persistent context and memory for AI coding agents across sessions and restarts. It is built as an open-source project for developers using AI coding agents. celeborn-code is commercial open source (BUSL-1.1). It runs on the web and the command line.
Behind celeborn-code is cloud-dancer-labs, and the product first shipped in 2026. Development happens publicly on GitHub with 15 commits in the last 90 days. Key capabilities include persistent memory, context management, and multi-agent kanban.
Latest indexed changes and source events
celeborn-code verified by the PulseGate indexer
Other apps tracked under the same category.