Review Design

Validate an experiment plan before execution using triage-first, fail-fast dimensional analysis with an adversarial red-team. Emits verdict (GO/REVISE/STOP), experiment_type, evaluation_dashboard, and revision_guidance.

5 stars 1 forksSoftware Development

Install

skills CLI (Claude Code, Cursor, Codex and more)
npx skills add TalonT-Org/AutoSkillit --skill review-design
Manual install
git clone --depth 1 https://github.com/TalonT-Org/AutoSkillit.git /tmp/AutoSkillit
mkdir -p ~/.claude/skills
cp -r /tmp/AutoSkillit/src/autoskillit/skills_extended/review-design ~/.claude/skills/review-design

Personal skills live in ~/.claude/skills. Use .claude/skills in a project folder to share a skill with your team.

From the SKILL.md

# Review Design Skill

Validate the quality and feasibility of an experiment plan before compute is spent.
Runs a triage-first, fail-fast multi-level analysis hierarchy with parallel subagents
and an adversarial red-team, then synthesizes a GO/REVISE/STOP verdict.

## When to Use

Use when the research recipe's `review_design` ingredient is `true` (the default). The
recipe calls this skill after `plan_experiment` to gate execution on a quality check.
This skill is bounded by `retries: 2` — on exhaustion the recipe proceeds with the
best available plan.

## Arguments

`/autoskillit:review-design {experiment_plan_path}`

- **experiment_plan_path** — Absolute path to the experiment plan file produced by
  `/autoskillit:plan-experiment`. Scan tokens after the skill name for the first
  path-like token (starts with `/`, `./`, or `.autoskillit/`).

## Critical Constraints

**NEVER:**
- Modify the plan file, any source code, or any file outside `{{AUTOSKILLIT_TEMP}}/review-design/`
- Halt the pipeline for a REVISE verdict — emit the verdict and let the recipe route
- Proceed to Level 2, 3, or 4 analysis when any Level 1 finding is classified as
  STRUCTURAL (halt at fail-fast gate). ADDRESSABLE L1 criticals continue L2-L4.
- Spawn SILENT (S) dimension subagents — they are not run and not mentioned in output
- Exit non-zero — GO, REVISE, and STOP are all normal outcomes (exit 0 in all cases)
Read the full skill on GitHub

Details

Skill name
review-design
Path
src/autoskillit/skills_extended/review-design/SKILL.md
License
MIT
Repo updated
8 November 2026

FAQ

What does the Review Design skill do?

Validate an experiment plan before execution using triage-first, fail-fast dimensional analysis with an adversarial red-team. Emits verdict (GO/REVISE/STOP), experiment_type, evaluation_dashboard, and revision_guidance.

How do I install the Review Design skill?

skills CLI (Claude Code, Cursor, Codex and more): npx skills add TalonT-Org/AutoSkillit --skill review-design. Skills go in ~/.claude/skills (personal) or .claude/skills (project), and Claude loads them automatically when a task matches the description.

Is Review Design popular?

It lives in TalonT-Org/AutoSkillit, which has 5 GitHub stars and 1 forks.

Related skills

Analyze Prs

5

TalonT-Org/AutoSkillit

Analyze all open PRs targeting a base branch — determine merge order, identify file overlaps, and tag each PR as simple or needs_check for complexity. Use at the start of a PR consolidation workflow.

Git & Code Review

Arch Lens C4 Container

5

TalonT-Org/AutoSkillit

Create C4 Container architecture diagram showing static structure, building blocks, and technology choices. Anatomical lens answering "How is it built?

Software Development

Arch Lens Concurrency

5

TalonT-Org/AutoSkillit

Create Concurrency architecture diagram showing parallel execution patterns, thread pools, synchronization, and barriers. Physiological lens answering "How does parallelism work?

Software Development

Update Architecture

5

TalonT-Org/AutoSkillit

Update architecture markdown files. Use when user says "update architecture", "update arch docs", "sync architecture", or mentions updating architecture documentation.

Software Development

Make Scenarios

5

TalonT-Org/AutoSkillit

Analyze a plan or codebase to generate scenarios in "Actor wants to..." format from a stated perspective. Use when user says "make scenarios", "generate scenarios", "jeopardy requirements", or wants to identify extension points and use cases.

Software Development

Update Specs

5

TalonT-Org/AutoSkillit

Add or update functional specification documents. Use when user says "update specs", "update specifications", "sync specs", or wants to maintain specification documents that describe what the software must do.

Software Development

Browse the most popular Claude skills, all software development skills, or write your own with the skills generator.