Skip to main content

Skillsync

Skillsync is a Mac app and CLI that gives AI coding agents persistent memory by automatically searching all past sessions before each prompt and delivering relevant recollections. It natively reads transcripts from Claude Code, Codex, Cursor, OpenCode, and other tools with zero configuration, performing on-device searches in under one second. Users can also trigger a global search overlay from any app or continue sessions across different agents.

Traction –/10
San Francisco, United States · HQ
31K+ followers
  • Developer Tools
  • Artificial Intelligence
  • Software Only
Updated 8 days ago

Funding

Raised to date

$500KRaised to date based on public sources. This may differ from the amount the company actually raised and is based only on what is publicly available on the internet.

Funding rounds are not available yet.

Founders

Founder details are not available yet.

Product

Problem

AI coding agents operate without memory of prior sessions, so developers frequently re-explain solved problems, repeat debugging loops, and lose context when switching between tools like Claude Code, Codex, and Cursor. Session logs are fragmented across multiple agent tools, making it impractical to manually search them before each prompt.

Solution

Skillsync provides a Mac app and CLI that automatically search all past agent sessions before each prompt and deliver relevant recollections to the active coding agent. It reads transcripts natively from Claude Code, Codex, Cursor, OpenCode, Amp, pi, Grok, Antigravity, and Hermes, requiring no indexing or setup. The system decides whether past work matters for the current prompt, staying out of the way when it does not, and reads only the sessions relevant to the current folder or scope. Search happens in under one second on-device, and users can also trigger a global search overlay from any app with a keyboard shortcut. Sessions can be continued in another agent, cropped, exported, or shared with teammates via link.

Target Audience

Primary customers are software developers and engineering teams who use AI coding agents daily and need persistent context across sessions, as well as organizations standardizing on multiple agent tools across local and remote development environments.

Features

  • Automatic pre-prompt search across all past sessions with relevance-based recall in under one second
  • Native session support for Claude Code, Codex, Cursor, OpenCode, Amp, pi, Grok, Antigravity, and Hermes
  • Zero-configuration local indexing with all session data kept on-device unless explicitly shared
  • Global search overlay (⌥⇧Space) accessible from any application with full-session preview and copy
  • Cross-agent session continuation, letting users resume a conversation in a different coding tool with full context
  • CLI (skl) for Linux, Windows, and remote servers supporting list, query, search, view, continue, crop, and export operations
  • Session sharing via link or by username/email, with recipients able to read the entire conversation
  • Skill library view showing each agent's capabilities, toggleable per agent, with linked sessions that used them
This profile is AI-generated and may contain inaccuracies.