Weekly refresh of PKA's Claude Code reference cache.
Fetches the canonical Claude Code docs (CLI reference, settings, hooks,
slash commands, sub-agents, skills, MCP, plugins, SDK, memory, IDE
integrations) plus the GitHub CHANGELOG.md, and writes a concatenated
markdown cache to Vault/memory/reference_claude_code.md in the PKA repo.
The git-sync task (runs every 15 min) commits and pushes the updated
cache to GitHub, where Lucienne's Mac picks it up on next pull.
Why: Lucienne should never answer Claude Code questions from training data — the CLI gains new flags faster than any model's cutoff. Having a fresh local cache lets sessions verify claims instantly instead of routing every question through WebFetch.
Schedule: Monday 07:00 UTC — early enough to be fresh for the week.
Script: ~/workspace/scripts/claude_code_docs_refresh.py Output: ~/workspace/PKA/Vault/memory/reference_claude_code.md
python3 /home/lucienne/workspace/scripts/claude_code_docs_refresh.py
claude directly. If the scheduler profile is GLM, Kimi, or MiniMax, prefer ~/workspace/PKA/scripts/claude-provider-run.sh so backup providers run in API-key-only mode with explicit settings/MCP context.
Format: minute hour day month weekday
| Started | Status | Duration | Log |
|---|---|---|---|
| 5d ago | completed | 4s | View log |
Mission Control is the board for your delegated work: requests come in, Luci coordinates the next step, and evidence stays visible for review.
Luci is your always-on assistant for routing, status updates, and follow-through. Operators can still open deeper evidence when needed.