Skill

plan-technical-review

Reviews implementation plans for simplicity, best practices, scope, and PR readiness; runs parallel agents and handles splitting large plans into parts.

code-quality

git-workflow

npx claudepluginhub verygoodopensource/very_good_claude_code_marketplace --plugin vgv-wingspan

Tool Access

This skill uses the workspace's default tool permissions.

Preview

**Plan file:** `$ARGUMENTS` (if empty, ask the user for the plan path or pick the most recent file under `docs/plan/`).

Supporting Assets

references/clear-context-handoff.mdreferences/extensive.mdreferences/minimal.mdreferences/standard.md

SKILL.md

Similar Skills

dev-plan-reviewer

Dispatches reviewer subagent to verify PLAN.md quality, task decomposition, and spec alignment before implementation phase. Handles chunking for large plans.

workflows

autoplan

Performs parallel multi-angle reviews of implementation plans from CEO, engineering, design, and devex perspectives. Merges scorecards, presents consolidated fix checklist for user selection, and applies approved changes.

claudekit

plan-exit-review

Interactively reviews implementation plans before coding, challenging scope creep, architecture, quality, tests, and performance with mandatory user checkpoints and opinionated recommendations.

1 tool

majestic-engineer

Stats

Stars59

Forks2

Last CommitMay 5, 2026

Actions

View Source View Plugin View on GitHub View README

Help us improve

Share bugs, ideas, or general feedback.

Plan technical review

Plan file: $ARGUMENTS (if empty, ask the user for the plan path or pick the most recent file under docs/plan/).

Run the following agents in parallel to conduct a comprehensive technical review of the plan at $ARGUMENTS. Pass the plan path to each agent:

@code-simplicity-review-agent: Review the plan for simplicity and clarity. Ensure the implementation is as straightforward as possible while still meeting all requirements.
@vgv-review-agent: Review the plan for adherence to Very Good Engineering practices and project conventions. Ensure the implementation follows our established patterns and conventions.
@plan-splitting-agent: Assess plan scope and recommend splitting into multiple PRs if the plan is too large for a single reviewable PR.

After all agents complete, if the plan-splitting-agent recommends a split:

Present the proposal to the developer via AskUserQuestion with options:
- Apply this split: generate separate plan files
- Keep as single PR: proceed without splitting
If approved, generate separate plan files:
- The skill (not the agent) generates the files
- Naming: docs/plan/YYYY-MM-DD-<type>-<original-slug>-part-N-plan.md
- Each file is a standalone plan following the same template and detail level as the original plan
- Each file includes all sections /build expects: title, type, acceptance criteria, tasks, file references. Use standard template by default; use minimal for simple parts or extensive for complex parts.
- Each file includes a ## Dependencies section noting which prior PR(s) must merge first
- Add a note at the top of the original plan file: > **Note:** This plan has been split into parts. See the `-part-N` files in this directory.

If the plan-splitting-agent reports no split needed: include the scope summary in the review output, no further action.

Handoff

When invoked directly by the user, use AskUserQuestion to present next steps after the review is complete:

Question: "Technical review complete! What would you like to do next?"

Options:

Clear context and build (Recommended): clear context for a fresh start, then build
Start building: execute the plan with /build
Refine the plan: improve the plan based on review findings
Done for now: review complete

If the user selects "Clear context and build" → Follow the clear context handoff for /build with the actual plan file path. Then stop.

When invoked by another skill (e.g., from /plan), return control to the caller after the review completes — do not present handoff options.