What problem does it solve? Years of AI-assistant history (ChatGPT, Claude, Perplexity exports and agent session logs) sit as unusable JSON blobs in downloads folders, making it impossible to answer questions like "when did I first discuss X" or trace how an idea evolved across past conversations. ## Core Features & Use Cases - Bulk Import Pipeline: Convert ChatGPT, Claude, and Perplexity exports plus agent session transcripts into one dated markdown page per conversation under conversations/, with mandatory secret/PII redaction before any page is written. - Validation & Fact Extraction: Validate every page against the native conversation parser and run cost-capped fact extraction via gbrain extract-conversation-facts. - Gap-Healing Archive: Detect missing conversations by diffing source dates against archived pages over a trailing window and backfill automatically, so no day is silently lost. - Retrieval & Tracing: Answer "when did I first discuss X" with dated slugs and verbatim quotes, and build idea-evolution timelines across the archive. - Use Case: You downloaded a ChatGPT export with 4,000 threads. The skill converts them into validated, redacted, date-prefixed pages, extracts facts, heals gaps, and lets you trace when a pricing idea first appeared. ## Quick Start Import my ChatGPT export conversations.json into the brain as dated conversation pages and tell me when I first discussed agent memory.
