30 Minutes Every Morning Lost to Re-Explaining Himself: The 20-Line Auto-Brief That Fixed It

Description and use

A 20‑line auto‑brief injected into Claude Code each morning eliminates repetitive explanations, saving over 30 minutes weekly.

Problem

Claude Code loses context between sessions, causing daily explanation round trips that waste 30 minutes and degrade precision of work.

Outcome

The automation recipe eliminates daily explanation round trips, saving over 30 minutes per week and ensuring Claude Code starts with accurate, up‑to‑date context.

Workflow

• 1. launchd triggers daily-brief.sh at 8:00 and 10:30 and on login
• 2. daily-brief.sh collects environment state: live probes, improvement log, hook latency, API cost, launchd status, git status of 10 repos, auto‑skill count, disk usage
• 3. Output is written to ~/.claude/logs/daily-brief-latest.md and other locations
• 4. cc-brief.sh runs on every Claude Code message submit
• 5. cc-brief.sh reads ccusage, settings, claude mcp list, improvements log, MEMORY.md, hot.md
• 6. cc-brief.sh injects the collected data into the system prompt as a brief
• 7. Claude Code starts the session with current state context, eliminating the need for re‑explanation

Evidence

The solution is automating context injection. Every time Claude starts a session, a 20–30 line summary describing the current state of the environment gets automatically inserted into the system prompt.

Source

This two-stage structure wiped out the morning "explanation round trip" entirely.

Source