Framework showdown: Superpowers, BMAD, SpecKit, GSD, OpenSpec.

TL;DR

Framework Comparison: Article by Rick Hightower compares five agentic coding frameworks with over 170,000 combined GitHub stars aiming to replace informal "vibe coding."[[1]](https://ai.plainenglish.io/the-great-framework-showdown-superpowers-vs-bmad-vs-speckit-vs-gsd-360983101c10)
Distinct Approaches: BMAD simulates enterprise teams, SpecKit uses gated specs, OpenSpec focuses on brownfield deltas, GSD isolates context in waves, Superpowers enforces test-driven development (TDD).[[1]](https://ai.plainenglish.io/the-great-framework-showdown-superpowers-vs-bmad-vs-speckit-vs-gsd-360983101c10)
Decision Aid: Provides a decision matrix and hybrid strategies to select tools without wasting time on mismatches.[[1]](https://ai.plainenglish.io/the-great-framework-showdown-superpowers-vs-bmad-vs-speckit-vs-gsd-360983101c10)

The story at a glance

Rick Hightower's practitioner analysis pits leading agentic coding frameworks against each other to guide developers picking structured AI-assisted development tools. The five compared are Superpowers, BMAD, SpecKit, OpenSpec, and GSD, each with unique bets on agent structure amid their rapid rise to over 170,000 GitHub stars combined. This review emerges as these tools explode in popularity by early 2026, replacing ad-hoc "vibe coding" in AI workflows.[[1]](https://ai.plainenglish.io/the-great-framework-showdown-superpowers-vs-bmad-vs-speckit-vs-gsd-360983101c10)

Agentic coding frameworks structure AI agents (like those in Claude Code or Cursor) for reliable software development, moving beyond chat-based trial-and-error.

Key points

Five frameworks hold over 170,000 combined GitHub stars, signaling a shift from "vibe coding" to disciplined processes.[[1]](https://ai.plainenglish.io/the-great-framework-showdown-superpowers-vs-bmad-vs-speckit-vs-gsd-360983101c10)
BMAD: Enterprise team simulation via multi-agent workflows for complex projects.[[1]](https://ai.plainenglish.io/the-great-framework-showdown-superpowers-vs-bmad-vs-speckit-vs-gsd-360983101c10)
SpecKit (GitHub): Gated specification process with CLI commands like `/specify` for spec-first development across 30+ agents.[[1]](https://ai.plainenglish.io/the-great-framework-showdown-superpowers-vs-bmad-vs-speckit-vs-gsd-360983101c10)[[2]](https://github.com/github/spec-kit)
OpenSpec: Brownfield-first delta specs for existing codebases, lightweight and tool-agnostic (20+ platforms).[[1]](https://ai.plainenglish.io/the-great-framework-showdown-superpowers-vs-bmad-vs-speckit-vs-gsd-360983101c10)[[3]](https://github.com/Fission-AI/OpenSpec)
GSD (Get Shit Done): Context-isolating wave parallelism for end-to-end execution in multi-phase projects.[[1]](https://ai.plainenglish.io/the-great-framework-showdown-superpowers-vs-bmad-vs-speckit-vs-gsd-360983101c10)[[4]](https://github.com/gsd-build/get-shit-done)
Superpowers: TDD-enforced discipline with composable skills like brainstorming and git worktrees for verified code.[[1]](https://ai.plainenglish.io/the-great-framework-showdown-superpowers-vs-bmad-vs-speckit-vs-gsd-360983101c10)[[5]](https://github.com/obra/superpowers)

Details and context

The article, paywalled on Medium, teases a decision matrix to match frameworks to use cases like greenfield vs. brownfield or solo vs. team development. It favors GSD and Superpowers based on author's experience, but dissects all for balanced choice.[[1]](https://ai.plainenglish.io/the-great-framework-showdown-superpowers-vs-bmad-vs-speckit-vs-gsd-360983101c10)

Each framework addresses AI agent flaws like context decay or skipped verification:

SpecKit and OpenSpec emphasize specs as "ground truth" in git.
BMAD scales to enterprise via role-based agents.
GSD waves prevent overload; Superpowers gates via TDD and reviews.

These tools integrate with agents like Claude Code, Cursor, Copilot; stars vary (Superpowers ~100k+, SpecKit ~70k, others lower), fueling 2026's spec-driven trend.[[6]](https://andrew.ooo/answers/best-agentic-coding-frameworks-2026)[[7]](https://medium.com/@richardhightower/agentic-coding-gsd-vs-spec-kit-vs-openspec-vs-taskmaster-ai-where-sdd-tools-diverge-0414dcb97e46)

Key quotes

"The vibe coding era is over. Five frameworks with 170,000+ combined GitHub stars are competing to define what replaces it. Each one makes a different bet on how much structure AI agents need."[[1]](https://ai.plainenglish.io/the-great-framework-showdown-superpowers-vs-bmad-vs-speckit-vs-gsd-360983101c10)

— Rick Hightower, article teaser.

Why it matters

Developers face weeks of trial-and-error picking AI tools; this comparison cuts that risk in a market where agentic frameworks drive productivity gains. Teams can match tools to needs—like TDD for solos (Superpowers) or specs for enterprises (SpecKit)—avoiding production bugs from unstructured AI code. Watch framework updates, benchmarks, and hybrids, as no single winner emerges yet and adoption speeds evolution.