Generates ranked maximalist alternatives to a proposed solution with calibrated complexity scores.
npx skills add https://github.com/Barbaros911/As-mine --skill overkill-barbaros911
Please help me install this Agent Skill. Skill: overkill Source: https://github.com/Barbaros911/As-mine/tree/main/.claude/skills/overkill Command: npx skills add https://github.com/Barbaros911/As-mine --skill overkill-barbaros911
--max filters to 🔥 7+ frontier options, --advanced switches to operator-focused comparison columns (ops burden, hiring difficulty, time to first commit), and --current uses web search to verify project health and refresh references. - Use Case: A staff engineer writing a tech-design document invokes overkill --advanced on a proposed in-process cache to survey and rule out options like CRDTs, consistent hashing, and Materialize with real adoption-cost framing. ## Quick Start Ask the AI to run overkill on your current solution, for example by saying "overkill this" after describing your pragmatic approach.High-intent search queries and answers about installing and using this skill.
Invoke overkill after describing your pragmatic solution. The skill restates your baseline, then proposes 3-6 ranked alternatives spanning data structures, algorithms, frameworks, patterns, and tooling, each with a complexity score and learning links.
--max restricts output to options at complexity 7/10 and above. --advanced switches the comparison table to operator-focused columns like ops burden and hiring difficulty. --current uses web search to verify project health and refresh references.
The scale runs from 1-2 (library swap, shippable in hours) through 5-6 (new runtime dependency with an SLO) and 7-8 (distributed-systems concerns requiring on-call) up to 9-10 (research-grade or specialist-hiring commitments). Scores are anchored so they stay comparable across invocations.
Do not use it when you want the simplest sufficient solution, an MVP, or a direct production recommendation. The skill explores the maximalist end of the design space, so presenting its output as a default recommendation would mislead.
Yes, it degrades gracefully. If web search is unavailable, the skill falls back to offline behavior with training-cutoff references and adds a one-line note that --current was requested but web access is not available.
causify-aiGenerate structured technical blog posts about machine learning libraries and techniques.
AIMFllysConvert requirements into justified architecture decisions with ADR documentation.
Agentic-Engineering-AgencyGuide pre-specification exploration with problem understanding, research, prototyping, and documentation.
matchaconeGuide developers toward minimalist, efficient coding solutions.