Stats
Actions
Tags
From mercury
Read and present an existing Mercury audit report. Use when the user wants to review, summarize, or query findings from a previously run audit without re-running the API call. Accepts a path to a report file or a project directory (auto-discovers all mercury_*_report.md files inside).
How this skill is triggered — by the user, by Claude, or both
Slash command
/mercury:mercury-reportThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Reads one or more existing `mercury_*_report.md` files and presents findings inline.
Reads one or more existing mercury_*_report.md files and presents findings inline.
.md file path, read that file directly.mercury_*_report.md files inside it:
find "<dir>" -maxdepth 2 -name "mercury_*_report.md" | sort
/mercury-audit first.For each report file found, extract:
```json … ```)Present in this format:
## Mercury Report — <report filename>
### Critical & High Findings
| # | Sev | Category | File | Symbol | Issue | Confidence |
|---|-----|----------|------|--------|-------|-----------|
...
### Medium Findings (count)
<collapsed list — file: issue>
### Top 5 to Fix First
1. ...
### Feasibility Verdict
<paragraph>
After presenting, offer:
/mercury-diff — compare this report against an older one/mercury-audit <dir> — re-run to get fresh findings/mercury-report [path_to_report_or_project_dir]
npx claudepluginhub mli-corp/mercury-audit --plugin mercuryCreates, edits, and optimizes skills for Claude Code, including drafting, evaluating with test prompts, iterating on performance, and improving skill descriptions for better triggering accuracy.