Identify core problems and risks in submitted code artifacts with evidence.
npx skills add https://github.com/DBvc/skills --skill dbx-linus-review
Please help me install this Agent Skill. Skill: dbx-linus-review Source: https://github.com/DBvc/skills/tree/main/skills/dbx-linus-review Command: npx skills add https://github.com/DBvc/skills --skill dbx-linus-review
This skill provides blunt, evidence-based technical critique of code changes, architecture plans, and implementation proposals, ensuring issues are surfaced with data-driven reasoning rather than vibes.
Provide a diff, code snippet, or architecture proposal and request a Linus-style strict review.
High-intent search queries and answers about installing and using this skill.
A Linus-style code review is a strict, pragmatic evaluation of artifacts that identifies core problems and articulates risks to users and systems. It scopes findings to data models, ownership, compatibility, and practicality using concrete evidence.
To get an evidence-based architecture review, provide a diff, code snippet, or architecture proposal. The review scopes to data models and compatibility, outputting concrete findings with suggested fixes and confidence levels for merge decisions.
Yes, you can use a strict technical evaluation to assess data model ownership and compatibility. It scopes the review to these specific areas, prioritizing real-world impact and providing data-driven reasoning rather than subjective vibes.
Yes, this strict review approach provides actionable outputs. Findings include concrete evidence, impacts, suggested fixes, and confidence levels, ensuring clear implications for your merge decisions.
The limitation of a Linus-style code review is its strict, blunt nature focused purely on technical critique. It prioritizes real-world impact and practicality, which may omit softer collaborative feedback typically found in standard code reviews.
hagajiVerify proposals, plans, and artifacts for alignment with intent.
telihaDetect specification contradictions and inconsistencies across markdown documents.
BetaSAnalyze code repositories read-only to provide evidence-based insights.
pacphiProvide unvarnished technical critique in Linus, Ramsay, and Bach modes.
