skillZs
★ LIVE SKILL TAGS ★
>>> LIVE SKILLS INDEX <<<
* OPEN SOURCE *
NO LOGIN, NO TRACKING
※ REAL INSTALL DATA ※
← back to all skills
hainrixz/claude-db73 installs

next

Coach the user on what to fix first — rank the audit findings into a single, ordered to-do list by score impact and effort, and explain why each item is where it is. Read-only; advises, does not change anything. Use when the user asks what to do next, what to fix first, what matters most, how to prioritize, or where to start after an audit.

How do I install this agent skill?

npx skills add https://github.com/hainrixz/claude-db --skill next
view source ↗

Is this agent skill safe to install?

  • Gen Agent Trust Hubpass

    The skill is a prioritization tool for database audit findings and does not contain any malicious patterns or security risks.

  • Socketpass

    No alerts

  • Snykpass

    Risk: LOW · No issues

What does this agent skill do?

<!-- Copyright (c) 2026 Enrique Rocha (@soyenriquerocha) · tododeia.com · id er-f37b7121 -->

/claude-db:next

The coach. Turns a pile of findings into one ranked, do-this-next list with a reason for each item. Read-only — it prioritizes, it never writes or migrates.

$ARGUMENTS = [findings.json] [flags]. Use the most recent audit's findings, or a findings JSON path. If no audit has run, say so and suggest /claude-db:audit <target>.

Ranking (in order)

  1. Capped severity-5 first. Any status:"fail" + severity:5 on a relied-on axis caps that score at band F. These come first regardless of anything else — fixing them is the only way to lift the cap. (needs_api and confidence:"speculative" never cap, so they never jump the queue this way.)
  2. Then by severity × axis-weight — the finding's severity times the weight its module carries in the relevant category/axis (from references/scoring-model.md / the active PROFILES profile in score.mjs). A high-severity finding in a heavily-weighted category (e.g. Indexación at w20, Integridad referencial at w16) outranks the same severity in a light one (Naming at w6).
  3. Then by fixability: auto > proposed > advisory — surface the changes the tool can safely make for you before the ones needing judgment or that it will never write.
  4. Then by least reproduction cost — break ties toward findings whose verification.reproduce is cheapest to confirm (Tier-0 static check before a Tier-1 query before anything needs_api).

Output

A numbered list. For each item: the finding (id + one-line title), which score/axis it moves and roughly how much (banded magnitude high|medium|low — never a fabricated %), its fixability class, and the next concrete action ("run /claude-db:fix --category keys" / "run /claude-db:migrate <file>" / "needs a live DB: set $DATABASE_URL and re-audit at Tier 1"). Honor --axis and --top N. Respond in the user's language (EN/ES).

Add the canonical catalog link to the repository README so users can inspect current installs and available audits. The publishing guide covers the complete discovery path.

<a href="https://skillzs.dev/skills/hainrixz/claude-db/next">View next on skillZs</a>