OvidCommunity·9 Agent Skills Included

paad

Spec pushback, code review, architecture audits, and TDD guardrails

Reviews specs, plans, branches, and architecture before and after code gets written. Catches contradictions, scope creep, bugs, security gaps, and accessibility barriers that fast code generation quietly ships. Runs multi-specialist reviews with verified findings, then guides test-first fixes so engineers stay in charge of every decision.

92|9

↗

npx skills add Ovid/paad --all -g -y

Available:

Tells the agent how the plugin's preview and shipped skill trees are organized, which versioning and promotion rules to enforce, and how to behave when working inside this repository.

All Skills in This Repository (9)

Pure Emerald Level Indicators

📦 In Repo

Ovid

agentic-architecture

Diagnose codebase architecture using coordinated multi-agent analysis and evidence-based reports.

agentic-a11y

Automate multi-agent accessibility audits across codebases with WCAG 2.2 AA guidance.

agentic-review

Automate multi-agent code review of a branch against main.

makefile

Create or update Makefiles with standard targets and user approval.

fix-architecture

Fixes architectural flaws from agentic-architecture reports via a test-first workflow.

alignment

Compare intent and action documents to surface requirement gaps and scope creep.

pushback

Critique specs, PRDs, and design plans for gaps and feasibility.

vibe

Enforce TDD red/green/refactor cycles for small code changes.

help

List available paad skills and show detailed help for a specific one.

Community

Basic

Frequently Asked Questions

FAQPage Schema

How to install paad?▼

Run `npx skills add Ovid/paad --all -g -y` in your terminal to install all skills globally.

What does paad do for code review?▼

The agentic-review skill dispatches six specialist reviewers over your branch diff, verifies each finding against the actual code, and ranks confirmed bugs by severity before you merge.

Can paad check a spec before building?▼

Yes. The pushback skill critiques a spec or PRD against your real codebase and git history, flagging contradictions, missing requirements, and infeasible items one at a time.

Does paad work with Cursor and other coding assistants?▼

Yes. It runs natively in Claude Code and installs into 70+ other assistants including Cursor, Codex, Copilot, and Gemini CLI through the standard skill format.

Can paad fix architecture problems it finds?▼

Yes. Run agentic-architecture to produce a verified findings report, then fix-architecture walks you through each flaw with a test-first workflow and your approval at every step.