logbook

Export verbatim Gmail, iMessage, and WhatsApp conversations into per-thread markdown archives.

1|1|Updated Apr 28, 2026
One-click install
npx skills add https://github.com/powerset-co/powerpacks --skill logbook-powerset-co
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: logbook
Source: https://github.com/powerset-co/powerpacks/tree/main/packs/ingestion/skills/logbook
Command: npx skills add https://github.com/powerset-co/powerpacks --skill logbook-powerset-co

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Reconstructing a complete, faithful record of every conversation with a set of people across email and messaging apps is tedious and error-prone, and synthesis tools discard the original text. This Skill downloads everything for a list of people and preserves it verbatim as a searchable local markdown archive. ## Core Features & Use Cases - Verbatim multi-channel export: Pulls every Gmail thread, iMessage DM, and WhatsApp DM/group into one markdown file per thread or chat, with ## YYYY sections and stable IDs in frontmatter. - Append-only incremental sync: Tracks per-channel watermarks in a manifest so later runs append only new messages without duplicating or overwriting. - Local and free: All reads are local SQLite queries (msgvault, wacli, chat.db) with no LLM, no network spend, and no message caps; groups are included by default. - Use Case: Given a CSV of founders with emails and phone numbers, build a complete raw archive of every email thread, text, and WhatsApp group with them, then keep it current with periodic syncs. ## Quick Start Use the logbook skill to build a raw logbook from my people CSV at contacts.csv, archiving every Gmail, iMessage, and WhatsApp conversation verbatim.

Frequently Asked Questions about logbook

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I export all my iMessage and WhatsApp conversations with specific people?▼

Provide a CSV of people (names, emails, phones) and run the logbook export, which pulls every Gmail thread, iMessage DM, and WhatsApp DM or group into markdown files under .powerpacks/logbook/. Groups are included by default; use --no-groups to limit to DMs.

How is logbook different from deep-context?▼

Logbook keeps verbatim message text in a raw archive with no message caps and groups on by default, while deep-context synthesizes facts and discards the text. Logbook is the inverse: it preserves everything rather than summarizing it.

Why are iMessage results empty when running logbook?▼

Reading chat.db requires Full Disk Access, which agent helper processes do not inherit from your terminal. Run bin/logbook directly in a terminal that has Full Disk Access granted, such as Ghostty, for the iMessage channel.

Does logbook send my messages anywhere or cost money?▼

No. Every read is a local SQLite query with no LLM, no network calls, and no per-message cost. Verbatim bodies are written only to the gitignored .powerpacks/logbook/ directory on your machine.

Can I update the archive later without re-downloading everything?▼

Yes. The sync command reads per-channel watermarks from manifest.json and appends only newer messages to existing files. It is append-only, never overwrites, and re-running with nothing new is a no-op.

What are the limitations of WhatsApp history export?▼

WhatsApp depth is bounded by what the primary phone still holds, since backfill uses on-demand sync from the phone. Full history is not guaranteed, though the deepen step runs multiple backfill rounds scoped to the CSV people's chats.