How do I install the Thermo Nuclear Code Quality Review skill?

Run npx skillmds@latest add devton/thermo-nuclear-code-quality-review in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is Thermo Nuclear Code Quality Review safe to use?

This skill has not completed SkillMD's automated safety review yet. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Thermo Nuclear Code Quality Review?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Thermo Nuclear Code Quality Review free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

Who published Thermo Nuclear Code Quality Review?

devton (@devton) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Thermo Nuclear Code Quality Review by devton

Name: Thermo Nuclear Code Quality Review
Author: devton

File contents

agentic-workflows-blueprint.workflow.thermo-nuclear-code-quality-review

Goal

Perform an unusually strict review focused on implementation quality, maintainability, abstraction quality, and codebase health.

Scope

Applies to: Any code change or pull request diff across the codebase.
Does not cover: Direct code modification or executing fixes (use thermo-fix).

Triggers

"Run thermo-nuclear review"
"/thermo-nuclear-code-quality-review"
"Deep code quality audit"
"Harsh maintainability review"

Inputs

diffScope: Files or commit range to review.
existingRootDoc: Main instruction file (AGENTS.md, CLAUDE.md, etc.).

Invariants

Be ambitious about structural simplification: Search for "code judo" moves that make implementation simpler, smaller, and cleaner.
1000 line limit: Do not let a PR push a file past 1k lines without strong justification.
No random spaghetti growth: Reject ad-hoc conditionals or one-off branches inserted into unrelated flows.
Clean design over working code: Do not approve merely because tests pass if the structure is messier.
No silent fallbacks: Ensure boundary cleanliness and explicit error/state contracts.

Procedure

Identify all changed files in diffScope.
Evaluate against the Non-Negotiable Additional Standards:
- Structural simplification & Code Judo opportunities.
- Decomposition of files approaching 1k lines.
- Elimination of ad-hoc conditionals or leaky abstractions.
- Type contract and boundary cleanliness.
Produce a Severity-Ranked Finding Table:
- Blocker: Structural regressions, 1k+ line file growth, severe spaghetti.
- High: Major maintainability or boundary issues.
- Medium: Moderate cleanup or abstraction improvements.
- Low: Informational nits.
Output specific, actionable remedies for every Blocker and High finding.

Outputs

Severity-ranked finding table appended to walkthrough.md or conversation output.
Final verdict: APPROVED or FINDINGS REMAIN.

Review gate

No unaddressed Blocker findings.
No unaddressed High findings.
Implementation preserves architecture contracts.

References

../../SKILL.md
../thermo-fix/SKILL.md
../radioactive/SKILL.md
Interactive HTML View

---name: thermo-nuclear-code-quality-reviewdescription: Run an extremely strict maintainability review for abstraction quality, giant files, and spaghetti-condition growth.---## agentic-workflows-blueprint.workflow.thermo-nuclear-code-quality-review### GoalPerform an unusually strict review focused on implementation quality, maintainability, abstraction quality, and codebase health.### Scope- **Applies to**: Any code change or pull request diff across the codebase.- **Does not cover**: Direct code modification or executing fixes (use `thermo-fix`).### Triggers- "Run thermo-nuclear review"- "/thermo-nuclear-code-quality-review"- "Deep code quality audit"- "Harsh maintainability review"### Inputs- `diffScope`: Files or commit range to review.- `existingRootDoc`: Main instruction file (`AGENTS.md`, `CLAUDE.md`, etc.).### Invariants1. **Be ambitious about structural simplification**: Search for "code judo" moves that make implementation simpler, smaller, and cleaner.2. **1000 line limit**: Do not let a PR push a file past 1k lines without strong justification.3. **No random spaghetti growth**: Reject ad-hoc conditionals or one-off branches inserted into unrelated flows.4. **Clean design over working code**: Do not approve merely because tests pass if the structure is messier.5. **No silent fallbacks**: Ensure boundary cleanliness and explicit error/state contracts.### Procedure1. Identify all changed files in `diffScope`.2. Evaluate against the **Non-Negotiable Additional Standards**: - Structural simplification & Code Judo opportunities. - Decomposition of files approaching 1k lines. - Elimination of ad-hoc conditionals or leaky abstractions. - Type contract and boundary cleanliness.3. Produce a **Severity-Ranked Finding Table**: - **Blocker**: Structural regressions, 1k+ line file growth, severe spaghetti. - **High**: Major maintainability or boundary issues. - **Medium**: Moderate cleanup or abstraction improvements. - **Low**: Informational nits.4. Output specific, actionable remedies for every Blocker and High finding.### Outputs- Severity-ranked finding table appended to `walkthrough.md` or conversation output.- Final verdict: **APPROVED** or **FINDINGS REMAIN**.### Review gate- No unaddressed Blocker findings.- No unaddressed High findings.- Implementation preserves architecture contracts.### References- `../../SKILL.md`- `../thermo-fix/SKILL.md`- `../radioactive/SKILL.md`- [Interactive HTML View](./README.html)

Frequently asked questions

Thermo Nuclear Code Quality Review