Reviews tool configurations for correctness.
Scanned 9/3/2026
Install to Claude Code
npx -y skills add serpro69/claude-starter-kit --skill before --agent claude-codeInstalls into .claude/skills of the current project.
Are you the author of Before?
Add the live security badge to your README — it updates automatically with every re-scan.
[](https://www.skillsdirectory.com/skills/serpro69-before-4b8775d2)More formats (shields.io, HTML) on the badges page.
---
name: review-tool
description: Reviews tool configurations for correctness.
---
# Review Tool
**Goal: Validate tool configuration files against the project schema.**
## Conventions
Follow standard review conventions.
## Validation Rules
You MUST check every configuration file for:
- Required fields present (`name`, `version`, `entrypoint`)
- No deprecated fields (`legacy_mode`, `compat_shim`)
- Version string matches semver format
## Error Handling
When a validation error is found:
1. Record the file path, field name, and violation type
2. Continue checking remaining files — do not stop on first error
3. Group errors by severity: blocking vs. warning
## Required Outputs
- [ ] Validation report written to `docs/reports/`
- [ ] Inline summary with pass/fail counts
- [ ] Blocking errors indexed under `kk:review-findings`
## Workflow
**Mandatory order — load rules before validating.** Read all configuration schemas before examining any target files.
Phases: load schemas → scan files → classify errors → write report → present summary.
Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.
No comments yet. Be the first to comment!
Adversarial multi-agent planning skill. Self-orchestrates 5 hostile category members (unspecified-low, unspecified-high, deep, ultrabrain, artistry) via team-mode for ruthless cross-critique debate, distills only the defensible insights, then MANDATORILY hands the distilled insight bundle to the `plan` agent for executable plan formalization. Use when planning needs maximum rigor and surfacing of weak assumptions, blind spots, and over-engineering. Triggers: 'hyperplan', 'hpp', '/hyperplan', ...