Available now.
31 AI handoff targets supported
Examples include: Aider, Amazon Q, Amp, Antigravity, ChatGPT, Claude Code, Claude Desktop, Claude.ai, Cline, Codex, Crush, Cursor, DeepSeek, Factory Droid, Gemini (web), Gemini CLI, Grok, Kilo Code, Kimi, Kiro, OpenClaw, OpenCode, Perplexity, Qwen Code, Roo Code, Trae, Devin, Windsurf, Z.ai (GLM), Replit, Lovable.
7 core surfaces
Dashboard, VS Code/Cursor extension, MCP server, CLI, desktop app, autosave hooks, local LLM failover.
Windows, macOS, and Linux builds
Desktop builds are available for all three operating systems under one lifetime license.
Metered local token savings
Measured on a real repo using Hivy context trimming before handoff.
Comprehensive test suite
Automated coverage across recovery, formatting, graph analysis, memory, and handoff workflows.
Per-agent formatting
Each target gets a handoff shaped for how it reads context, not one generic blob.
Dependency graph
Impact radius and god-node detection warn the next AI what is dangerous to touch.
SHA-256 audit trail
Hash-chained log of every handoff - tamper-detecting and fully local.
Persistent project memory
Gets richer over time, so mature projects recover closer to the original session.
Session Vault
Always-on local recovery history keeps surviving conversation context available after crashes, closes, outages, or tool bugs.
4-tier context system
Session, task, project, or everything. ShardStitch selects the right depth for the next tool instead of dumping one oversized blob.
Integrity Engine
Fresh scans reconcile stale memory, changed files, and dependency risk before a handoff is generated.
Hivy on-device layer
Local models trim noisy context, catch drift, and route easy work away from frontier tokens.
49 MCP tools
Callable from Claude Code, Cursor, and Devin when you want the AI to scan, recall, hand off, run council, or request recovery context directly.