ponytail

Enforces minimal code solutions by prioritizing stdlib, native features, and shortest diffs.

Updated May 24, 2026
One-click install
npx skills add https://github.com/MWest2020/skill-forge --skill ponytail-mwest2020
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: ponytail
Source: https://github.com/MWest2020/skill-forge/tree/main/skills/ponytail
Command: npx skills add https://github.com/MWest2020/skill-forge --skill ponytail-mwest2020

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Codebases accumulate over-engineering: unnecessary abstractions, premature dependencies, boilerplate scaffolding, and speculative features. This Skill forces the laziest solution that actually works, channeling a senior developer who questions whether code needs to exist at all before writing it. ## Core Features & Use Cases - The Ladder: A six-rung decision hierarchy that stops at the first working solution, from YAGNI elimination through stdlib, native platform features, existing dependencies, one-liners, and only then minimal custom code. - Intensity Levels: Three modes (lite, full, ultra) controlling how aggressively minimalism is enforced, switchable mid-session with /ponytail lite|full|ultra. - Guardrails: Never simplifies away input validation, security, error handling, or accessibility; leaves one runnable self-check for non-trivial logic; marks deliberate shortcuts with ponytail: comments naming the ceiling and upgrade path. - Use Case: When asked to add a cache for API responses, instead of building a custom TTL cache class, it applies @lru_cache(maxsize=1000) and notes in one line what was skipped and when to revisit it. ## Quick Start Ask the assistant to solve your next coding task in ponytail mode, for example by saying use ponytail to add caching to this API client.

Frequently Asked Questions about ponytail

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I avoid over-engineering code with an AI assistant?

Activate ponytail mode and the assistant applies a decision ladder: question whether the feature is needed, prefer stdlib and native platform features, reuse installed dependencies, and write the shortest working diff. It ships minimal code with a one-line note on what was skipped.

What are the ponytail intensity levels lite, full, and ultra?

Lite builds what is asked but names a lazier alternative in one line. Full (the default) enforces the ladder with stdlib-first and shortest diffs. Ultra is YAGNI-extremist, preferring deletion over addition and challenging the requirement itself.

Does minimal code generation skip tests and error handling?

No. Input validation, security measures, error handling preventing data loss, and accessibility are never simplified away. Non-trivial logic gets one small runnable self-check, such as an assert-based __main__ block, without test frameworks or fixtures.

How do I turn off ponytail mode or switch levels?

Say "stop ponytail" or "normal mode" to revert to standard behavior. Switch intensity mid-session with /ponytail lite, /ponytail full, or /ponytail ultra. The chosen level persists until changed or the session ends.

When should I not use a minimalism-first coding approach?

Avoid it when requirements genuinely demand scalability headroom, when the user explicitly requests a full implementation, or at trust boundaries needing thorough validation. Ponytail itself builds the full version without re-arguing once the user insists.