vpaivag
01Plugin Claude Code
vpaivag's personal Claude Code plugins.
Plugin Claude Code
vpaivag's personal Claude Code plugins.
Plugin Claude Code
The BBQ flow (Blueprint → Build → QA) — a planning pipeline with blind QA and a visual review surface, plus adversarial review and a CLAUDE.md refactor utility.
Settings file Claude Code
Agent settings configuring enabledPlugins.
Instructions file
Instructions for vpaivag/skills, covering skills, hard rules and layout.
Agent
Split-context red/blue critic for a single plan proposal. Spawned by /blueprint during the design path — receives exactly one proposal plus the task constraints, never the codebase or sibling proposals, and returns only the attacks that survive its own repair attempt. Use one critic per proposal.
Agent
Blind QA-plan author. Spawned by /blueprint immediately after context.md is written — reads ONLY context.md and derives behavioral verification criteria from intent alone, before and without ever seeing any plan, proposal, or code. Writes qa-plan.md into the suite dir. Re-run whenever context.md's Task or Expected…
Agent
Renders a plan suite into visual.html — the human review-and-approval surface. Spawned by /blueprint after plan.md and qa-plan.md exist; also invocable directly ("render a visual plan for "). Reads the whole suite, scales depth to the plan, authors block markup against the plugin's block library, and regenerates the…
Skill Claude CodeCodex
Adversarial code review of a diff using split-context subagents — separate reviewer agents each get ONLY the diff and are told to assume the code is wrong and find the reasons it breaks, findings are then verified against the real code to drop false positives, and reported (fixes applied only on request). Use this…
Skill Claude CodeCodex
Front-door planning skill that takes a coding task from stated intent to an approved, executable plan suite. Gathers context (restate → explore fan-out → batched Q&A → persist context.md), gates on whether a genuine design choice exists, and scales accordingly — mechanical tasks get a direct plan, design tasks get 2–3…
Skill Claude CodeCodex
Refactor, audit, or bootstrap a repository's CLAUDE.md so it stays lean and acts as an index of pointers to /docs/ files instead of bloating session context. Use this skill whenever the user mentions their CLAUDE.md being too long, bloated, eating context, "collapsing context", needing cleanup, or asks to set up /…
Skill Claude CodeCodex
Guided onboarding for newcomers to a codebase. Three phases — gate on CLAUDE.md quality, run a conversational tour driven by CLAUDE.md and the docs it points at, then write one contrived hands-on task to .claude/onboarding-task.md for the newcomer to implement themselves on a sandbox branch. Read-only on the codebase…
Skill Claude CodeCodex
Performs thorough, expert-level pull request reviews and returns a structured report with issues categorized by severity and code snippets. Use whenever the user asks to review a PR, look at a pull request, do a code review, check a branch before merge, evaluate proposed changes, or asks "what do you think of.
Skill Claude CodeCodex
Verify a suite's behavioral intent by executing qa-plan.md in a plan-blind session. Reads ONLY qa-plan.md (and context.md) — never plan.md, visual.html, or the diff — then verifies each B-n/R-n criterion against the running code. Writes real test files when the repo has a test setup; otherwise performs manual…
Skill Claude CodeCodex
Configure where plan suites live and whether the plans directory is gitignored. Writes .claude/plans-config.json with plansDir (default .claude/plans) and gitignore (default true) keys. The planning skills (/blueprint, /build, /qa) read this config to resolve the plans directory, falling back to .claude/plans when the…