← ClaudeAtlas

optimiselisted

Analyse hooks daemon configuration and recommend improvements across Safety, Stop Quality, Plan Workflow, Code Quality, and Daemon Settings — the canonical config-optimisation step
LongTermSupport/fedora-desktop · ★ 6 · DevOps & Infrastructure · score 74
Install: claude install-skill LongTermSupport/fedora-desktop
# /optimise - Configuration Optimiser Skill (canonical config-optimisation step) ## Description This is the formalised "enable all relevant handlers and ensure optimal configuration" step (Plan 00308) — the same step whether run manually, automatically at the end of `/hooks-daemon upgrade`, or as the closing step of LLM-INSTALL.md/LLM-UPDATE.md. There is no separate `config-optimisation` command; `/optimise` IS it. Analyse the current hooks daemon configuration against the project's profile (languages, tests, CI, plans) and produce a scored report across five key areas. Also compares the project's config against `CLAUDE/UPGRADES/config-changes/` manifests to surface capabilities introduced since the last recorded run, and can apply recommendations automatically. Every run (report-only or apply) records itself via `bin/hooks-daemon record-config-optimisation-run`, which silences the `config_optimisation_reminder` SessionStart advisory until the next upgrade. ## Usage ```claude-code # Run full analysis and show scored report /optimise ``` No arguments — the skill profiles the project automatically. ## What It Checks The skill analyses five areas, scoring each PASS / WARN / FAIL: 1. **Safety** — Critical blocking handlers (destructive_git, sed_blocker, security_antipattern, etc.) 2. **Stop Quality** — Handlers that prevent poor stopping behaviour (auto_continue_stop, plus the nitpick.hedging_language and nitpick.dismissive_language detectors) 3. **Plan Workflow** — Plan