Contextify 1.0.6 is now available on the Mac App Store and as a direct download. This release brings three significant improvements.
Total Recall: Your AI Can Search Your History
This is the power feature of 1.0.6.
Contextify now includes a CLI tool that Claude Code can use to search your conversation history. Ask your AI:
- "What did we decide about the database schema last week?"
- "Find where we discussed authentication approaches"
- "What was that solution for the API rate limiting issue?"
Your AI can now reference past decisions, revisit previous solutions, and maintain context across sessions.
The feature consists of a CLI tool (contextify-query), a skill (/total-recall), and a researcher agent (contextify-researcher) that knows how to search effectively.
How to Enable Total Recall
DMG users: Go to Settings > CLI and click "Install/Repair CLI." The CLI and plugin install automatically.
App Store users: Due to sandbox restrictions, install the CLI via Homebrew:
brew tap peterpym/contextify
brew install contextify-queryAfter installation, restart Claude Code. The skill registers automatically and you'll have access to /total-recall.
macOS 15 (Sequoia) Support
The most requested feature: Contextify now works on macOS 15.
When I launched Contextify, it required macOS 26 (Tahoe) because it relies on Apple Intelligence for local LLM summaries. Many users wanted to try the app but weren't ready to upgrade to Tahoe.
Version 1.0.6 introduces Lite Mode for macOS 15:
- Timeline monitoring works fully
- Search across all your sessions works fully
- Transcript backup and organization works fully
- AI summaries are disabled (they require Apple Intelligence on macOS 26)
Start building your searchable history now. When you eventually upgrade to Tahoe, summaries will generate automatically.
Improved Agent Sidechain Capture
When Claude Code uses the Task tool or spawns sub-agents, those conversations are now captured and displayed in your timeline with visual indicators.
Previously, you'd only see the main conversation thread. Now you see the full picture of multi-agent work, including what each sub-agent did and how it contributed to the result.
Bug Fixes and Performance
- Improved Codex session discovery
- Faster historical transcript ingestion
- Various stability improvements
Questions or feedback? Email support or open an issue on GitHub.