auto-optimize

Solid

Analyze a target project, classify findings, and implement at most one user-confirmed optimization with focused verification.

AI & Automation 41 stars 7 forks Updated today MIT

Install

View on GitHub

Quality Score: 82/100

Stars 20%
54
Recency 20%
100
Frontmatter 20%
70
Documentation 15%
100
Issue Health 10%
50
License 10%
100
Description 5%
100

Skill Content

# Auto-Optimize: Bounded optimization process Use VibeGuard to assess a project and deliver one bounded, verified optimization. ## When to Activate - The user asks for repository optimization discovery and implementation. - A project needs a rotating scan across security, quality, reliability, performance, and maintainability. - Existing guard or audit findings should be converted into prioritized implementation work. - The task needs an optimization report before deciding which fixes to execute. - The user explicitly asks for optimization across quality, reliability, security, or DX dimensions. ## Red Flags - The workflow starts without a bounded target or explicit authorization to implement. - More than one writable session operates on the repository. - A finding is repaired before it has evidence, classification, and a verification command. ## Checklist - [ ] Confirm target, authorization, and done-when before execution. - [ ] Classify every finding as FIX, SKIP, or DEFER with evidence. - [ ] Run VibeGuard deterministic checks before and after implemented fixes. ## Execution Boundary Start only when the user authorizes implementation and the optimization target is bounded. Keep one writable session for the repository and one short issue-sized batch per run. Do not build coordinator/reviewer lanes by default. If the user explicitly requests delegation, follow [`workflows/references/delegation-contract.md`](../references/delegation-contract.md). ## Core principles...

Details

Author
majiayu000
Repository
majiayu000/vibeguard
Created
6 months ago
Last Updated
today
Language
Shell
License
MIT

Similar Skills

Semantically similar based on skill content — not just same category