scribe-cli

name

scribe-cli

description

Set up and operate the scribe CLI — install/init personal or team KBs, diagnose with doctor/status, run sync/extraction, manage cron and Full Disk Access, and pick the right command for a maintenance goal. Use when the user wants to install or run scribe, asks which scribe command or init/team flags to use, says their KB is stale/empty or extraction/sync stopped, or is reading `scribe doctor`/`scribe status` output. For authoring KB content use scribe-kb; for the `scribe lint` content-quality queue use scribe-kb-tidy.

scribe CLI — operations agent skill

scribe is a single-binary Go CLI that runs a knowledge-base pipeline: discover repos → extract knowledge → mine agent sessions → absorb URLs → lint → reindex with qmd. It normally runs unattended on cron. This skill is for driving the CLI by hand — diagnosing health, running a step manually, and choosing the right command for a goal.

Two sibling skills own the other halves and this skill routes to them:

• scribe-kb — authoring/searching KB content (articles, drop files, qmd).
• scribe-kb-tidy — working the scribe lint content-quality queue (split bloated, archive rolling, merge thin/self-named-dirs).

Installation or bootstrap requests

For zero-to-running setup, read the public, prompt-compatible runbook first: https://getscribe.dev/setup.md. It has separate command sequences for a personal Anthropic KB, local Ollama, a hosted OpenAI-compatible provider, the team owner, each team member, a second KB on one machine, and Linux cron. Follow one profile rather than mixing flags.

Key distinctions the short command map cannot express:

• Fresh/default KBs use scribe init --path <kb> --bind; --bind is what installs the machine-global Claude/Codex/Amp handshake.
• Only the team owner uses --team, , , and . Members clone, into the checkout, then run .