Assistants

stratless reads your coding assistant’s own history and loads a profile back into it. The core is the same for every tool: read your sessions, record what you did in each exchange, write a HUMAN.md, load it. Only the wiring differs per assistant, where the transcripts live, how the profile loads, and what cleanup you have to guard against.

It rolls out one tool at a time.

Supported today

Claude Code

• Reads ~/.claude/projects: JSONL transcripts, one per session.
• Loads the profile into ~/.claude/HUMAN.md, imported by your ~/.claude/CLAUDE.md, so it loads at the start of every session.
• Borrows claude -p to read your history and write the profile, on your own subscription.
• The reaper: Claude Code deletes transcripts after 30 days, per file. stratless init stops that (it sets cleanupPeriodDays to 3650) and archives everything it can still reach. This is the one thing you cannot do later.

Next: the CLIs

Aider, Gemini CLI, Codex, Cline, and Copilot all keep local session logs and expose a scriptable command to borrow. The architecture is one core and thin readers, so these come next: the same HUMAN.md, a per-tool reader and load path.

Later: the GUIs

Cursor, Zed, and Windsurf come after. Their histories are undocumented database blobs rather than clean logs, so reading them is messier and more fragile. HUMAN.md still rides out to them once the reader is solid.

Want yours supported?

If your assistant keeps a local, readable history, it can work. Open an issue and tell us which one.