global-review-looplisted

Mine ALL of your local Claude history across EVERY project (both corpora — Claude Code CLI and Cowork local-agent-mode) for operator friction that recurs across MULTIPLE projects, then propose GLOBAL ~/.claude leverage (a CLAUDE.md rule, a new/edited global skill, a memory, or a settings hook) to fix it — each proposal reconciled against what already ships globally, self-validated by an adversarial /review-loop --mode claim loop, and recorded in a cross-run ledger so nothing is re-proposed, dismissed-twice, or shipped on its own say-so. DISTINCT FROM: signal-scan (one repo → that repo's .claude/ config), command-retro (Command's own logs → Command's queue), and transcript-analysis (ONE project, Cowork-only, proposes that project's CLAUDE.md candidates). This skill is the GLOBAL, cross-project layer: it only surfaces friction proven to recur across >=N DISTINCT projects (counted by cwd-PATH attribution only) and only ever proposes changes to ~/.claude. Invoke as /global-review-loop. Triggers: "what global rule
BryceEWatson/claude-global-skills · ★ 0 · AI & Automation · score 75

Install: claude install-skill BryceEWatson/claude-global-skills

# /global-review-loop The **cross-project, global self-improvement pass.** It reads how you actually work across your WHOLE fleet — every project, both local Claude corpora — finds the friction that recurs in **two or more different projects**, and proposes the smallest **global `~/.claude`** change that would fix it. It PROPOSES only; it has **no write tools** and cannot edit `~/.claude` itself (see Apply boundary). It is one member of a deliberately-partitioned family — know which one you want before you run it. The differentiator is **scope + the cross-project recurrence gate**, not "the others don't propose" (two of them do): | Skill | Unit of work | Reads | Proposes changes to | Promotion bar | |---|---|---|---|---| | `/signal-scan <repo>` | ONE repo | that repo's CLI history | that **repo's** `.claude/` (proposes) | recurs within one repo | | `/command-retro` | Command itself | Command's own logs | Command's work queue (proposes) | Command-only friction | | `transcript-analysis` | ONE project | that project's **Cowork** transcripts | that **project's** `CLAUDE.md` (proposes per-project candidates) | recurs across that project's sessions | | **`/global-review-loop`** | the **whole fleet** | **ALL projects, BOTH corpora** | the **global `~/.claude/`** (proposes) | recurs across **>=N DISTINCT projects** (path-attributed) | | `/chat-history-search` | read-only | anything | nothing (search/inventory) | n/a | **Why global-review-loop is not a duplicate of transcript-anal