From purlin
Shows rule coverage dashboard with feature table, summary, and optional report link. Useful for tracking proof status across all features.
How this skill is triggered — by the user, by Claude, or both
Slash command
/purlin:statusThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Show rule coverage across all features. Always outputs a consistent table followed by a summary line and dashboard link.
Show rule coverage across all features. Always outputs a consistent table followed by a summary line and dashboard link.
purlin:status Show all features
purlin:status --role <role> Filter by role (pm, dev, qa)
sync_status(role: <from argument, optional>)
Always display a table with three columns: Feature, Coverage, and Status. Every feature and anchor must appear in this table, sorted by status (PARTIAL first, then PASSING, then VERIFIED). Anchors are labeled with (anchor) after the name.
Feature Coverage Status
─────────────────────────────────────────────────────────
purlin_report 12/41 PARTIAL
schema_spec_format (anchor) 7/8 PARTIAL
sync_status 38/41 PARTIAL
config_engine 15/15 PASSING
dashboard_visual (anchor) 11/11 PASSING
e2e_audit_cache_pipeline 15/15 PASSING
skill_find 4/4 VERIFIED
skill_build 7/7 VERIFIED
security_no_dangerous_patterns (anchor) 5/5 VERIFIED
Coverage format is proved/total rules. The table must include every feature and anchor — never omit rows or collapse them into a summary.
Do NOT print the per-feature detail breakdown (individual RULE/PROOF lines and → directives) in the table output. The table is a dashboard overview. If the user wants detail on a specific feature, they can use purlin:find <name>.
After the table, print a one-line summary:
Summary: 42 features | 10 VERIFIED | 20 PASSING | 6 PARTIAL | 3 FAILING | 3 UNTESTED
If purlin-report.html exists at the project root, print the clickable link:
Dashboard: file://<absolute-path-to-project>/purlin-report.html
Check for the file with a glob or ls before printing. If the file does not exist, skip this line.
| Status | Meaning |
|---|---|
| VERIFIED | All rules proved + verification receipt matches |
| PASSING | All rules proved, no receipt yet |
| PARTIAL | Some rules proved, none failing — more tests needed |
| FAILING | Any proof has status FAIL |
| UNTESTED | No proofs at all |
The progression is: UNTESTED → PARTIAL → PASSING → VERIFIED.
Provides behavioral guidelines to reduce common LLM coding mistakes, focusing on simplicity, surgical changes, assumption surfacing, and verifiable success criteria.
Searches, retrieves, and installs Agent Skills from prompts.chat registry using MCP tools like search_skills and get_skill. Activates for finding skills, browsing catalogs, or extending Claude.
Creates, edits, and optimizes skills for Claude Code, including drafting, evaluating with test prompts, iterating on performance, and improving skill descriptions for better triggering accuracy.
npx claudepluginhub rlabarca/purlin --plugin purlin