What problem does it solve? When a user references older or parent conversations that are no longer in active memory, there is no built-in way to recall what was said. This Skill provides ready-to-use jq and ripgrep commands to search, filter, and analyze complete conversation history stored in session JSONL files. ## Core Features & Use Cases - Full-Text Session Search: Search across all session transcripts for keywords or phrases using ripgrep, or filter by role (user, assistant, toolResult) with jq. - Usage and Cost Analytics: Compute per-session and daily cost totals, message counts, and token usage from the message.usage.cost.total fields. - Tool Usage Breakdown: Aggregate which tools were called and how often within any session transcript. - Use Case: A user asks "what did we decide about the database schema last Tuesday?" You locate sessions from that date, extract the user and assistant text messages, and search for the relevant discussion. ## Quick Start Search my session logs for any past conversation where we discussed database migration and summarize what was decided.