EffortlessMetrics

Cluster Triage — Development skill for Claude Code

Development community

Ensemble-curator step 6 — for a detected cluster, pick winner(s) by file-path, extract edge cases from losers, close dupes with cross-ref.

How to install Cluster Triage

Installs to ~/.claude/commands/effortlessmetrics-perl-lsp-cluster-triage.md

Terminal
mkdir -p ~/.claude/commands && curl -fsSL https://raw.githubusercontent.com/EffortlessMetrics/perl-lsp/HEAD/.claude/commands/cluster-triage.md -o ~/.claude/commands/effortlessmetrics-perl-lsp-cluster-triage.md

Restart Claude Code, or start a new session, for it to be picked up.

What Cluster Triage does


description: Ensemble-curator step 6 — for a detected cluster, pick winner(s) by file-path, extract edge cases from losers, close dupes with cross-ref

Cluster Triage

Given a cluster detected by `/ensemble-detect`, pick winner(s) and consolidate.

See `docs/articles/BROAD_SCOPE_LAYER_DIVERSITY.md` for the underlying file-path-over-title-triage principle.

The rule

**File-path first, title second.** Two PRs with similar titles touching the same file + same function = duplicates. Di

Alternatives in Development

  • Analyze Cluster Postmortem — Postmortem analysis of cluster failures → OPTIMIZE PROMPTS to work for EVERY use case 1.7k ★
  • HERO Anti OverDefense — HERO = Hashing · Edge cases · Rubrics · Overbuild — the four shapes coding agents over-defend in 333 ★
  • Eval Author — Draft a new eval case (eval/cases/ .json) + scaffold its golden folder, golden_pending until captured on the V 136 ★

Full documentation available on GitHub

View Source Repository