← ClaudeAtlas

stapler-squad-ruleslisted

Use when working in the stapler-squad repo to manage Claude Code approval rules — reviewing analytics for coverage gaps, adding or retiring rules, debugging why a session generates excessive manual review requests, or editing seed rules in pkg/classifier/classifier.go.
tstapler/stapler-squad · ★ 1 · AI & Automation · score 64
Install: claude install-skill tstapler/stapler-squad
# Stapler Squad Rules Engine Guides you through reading the approval analytics, identifying coverage gaps, and adding or tuning rules in the stapler-squad rules engine. ## When to Use This Skill - Reviewing analytics to understand what Claude is doing - Identifying gaps (commands not covered by any rule) - Deciding whether to add, modify, or retire an auto-approval rule - Investigating why a session is generating lots of manual review requests ## Accessing the Analytics ### Option A: UI (quick overview) Navigate to **http://localhost:8543/rules** (run `make restart-web` if the server is not running). Two panels: 1. **Approval Rules Panel** (top) — active rules with trigger counts 2. **Approval Analytics Panel** (bottom) — time-series data, coverage gaps, program breakdowns Use the **7 / 14 / 30 / 90 day** window selector. **Note:** if the panel shows "Loading analytics…" with 0 decisions but you know data exists, use Option B instead — the UI reads from the SQLite DB while the JSONL migration may still be in progress for recent data. ### Option B: REST API (authoritative, 90-day window) ```bash curl -s -X POST "http://localhost:8543/api/session.v1.SessionService/GetApprovalAnalytics" \ -H "Content-Type: application/json" \ -H "Connect-Protocol-Version: 1" \ -d '{"windowDays": 90}' -o /tmp/approval_analytics_api.json # Parse key metrics python3 - <<'EOF' import json, collections with open("/tmp/approval_analytics_api.json") as f: s = json.load(f)["summary"