← Back to the catalog Constrained optimization skill for maximizing or minimizing objectives under budgets, capacities, safety limits, equality/inequality constraints, and multi-objective trade-offs. Provides model normalization, ALM/ADMM/KKT routing, infeasibility diagnosis, shadow-price interpretation, capability-aware no-tool fallback, and prompt-injection-resistant execution.
View on GitHub ↗ Copy repo URL Copy SKILL.md link License: MIT /plugin marketplace add Sliky1/lagrangian-skills The exact command may vary by repository. Check the README on GitHub.
For the skill author
Shows your skill is listed on Skillteca, generates a backlink and trackable traffic.
Markdown HTML
[](https://www.skillteca.com.br/skills/lagrangian?utm_source=badge&utm_medium=readme&utm_campaign=badge) Copy snippet Resources to assist in writing various internal communications, adhering to company-preferred formats. Claude should utilize this skill for status reports, leadership updates, newsletters, FAQs, and other internal documents.
DevOps e Infra by anthropics
Monitors a pull request or review cycle until it is ready to merge. This skill is used to track PR comments, reviews, and CI status until all actionable issues are resolved.
DevOps e Infra #ai by thedotmack
Execute a phased implementation plan using subagents. Use when asked to execute, run, or carry out a plan — especially one created by make-plan.
DevOps e Infra #ai by thedotmack
Token-optimized structural code search using tree-sitter AST parsing. Use this instead of reading full files when you need to understand code structure, find functions, or explore a codebase efficiently.
DevOps e Infra #ai by thedotmack
Category alert
One short email with only the new DevOps e Infra skills. 4 minutes of reading, no spam, unsubscribe with one click.
You confirm your email on the first send. No spam. Unsubscribe with one click.
Lagrangian Skill — v1.0.0
Stable release | measured 96.78% on included benchmark summary | no fabrication | reproducible scaffolding included
0. Scope
支持: convex_qp | smooth_nlp | non_convex_nlp | distributed_admm | safe_rl_constraints | multi_objective | softenable_logic_constraints | mixed_bayes_opt_handoff
有限支持: OR/条件逻辑→smooth approximation或case split;若必须精确整数/二进制求解→OUT_OF_SCOPE
不支持: 纯贝叶斯推断 | 纯统计检验 | 精确MIP/整数规划 | 缺少关键参数的数值求解 | 未授权外部代码/网络执行
默认输出: STANDARD。用户说“只要答案/数字”→MINIMAL;用户说“展开/推导/
[Description truncada. Veja o README completo no GitHub.]
Read full description↓
Comments · No comments No comments yet. Be the first.