bmad-code-org/BMAD-METHOD

Breakthrough Method for Agile Ai Driven Development

JavaScript48148 ★AI Agent FrameworksGitHub ↗
Quality: solid 19/24
PAI: integrate 0.5
Appraised: 2026-05-27 current
Contents

Overview

Verdict

Rating Summary
Quality solid (19/24) Actively maintained at v6.8.0 with extraordinary adoption, strong engineering discipline, and thorough documentation; docked only for NOASSERTION license metadata and absent CI badge.
PAI Relevance integrate (0.50) Scale-adaptive planning patterns and 34-workflow lifecycle are worth adopting into PAI's Loop and Agents skills; bunx bmad-method install provides a concrete subprocess integration surface for structured project scaffolding.

INTEGRATE by formula; the value is workflow-pattern adoption rather than library-level code integration — BMAD's structured development lifecycle and cost-optimization approach (flat-rate web LLM for planning, metered IDE tokens only for implementation) are the primary integration surfaces.

Quality Assessment

19/24 — maintained / adequately-documented / high-discipline

Health: 6/8 (maintained)

Failed:

Passed:

Documentation: 6/8 (adequately-documented)

Failed:

Passed:

Engineering Signals: 7/8 (high-discipline)

Failed:

Passed:

PAI Relevance

Dimension Score Assessment
Harvest Value 1 BMAD's scale-adaptive planning-depth algorithm (auto-adjusts from bug-fix to enterprise scope) and its 34-workflow structured lifecycle offer concrete design patterns worth studying for PAI's Loop skill (iterative Algorithm cycles) and Agents skill (persona composition); the web-bundle cost model — flat-rate web LLM for planning work, metered IDE tokens only for implementation — is a novel cost-management pattern with no equivalent in the current PAI Capability Manifest.
Integration Readiness 1 Primary language is JavaScript rather than TypeScript, but ships a clean CLI distributable as bunx bmad-method install; PAI could invoke it as a subprocess for project initialization or adapt its YAML/markdown workflow artifacts directly into PAI skills with moderate adapter work; not a bun add drop-in.
Overlap Risk 1 Partial overlap with PAI's Agents skill (agent composition + voices), Delegation skill (parallel work routing), and Loop skill (iterative Algorithm cycles); BMAD's specific domain — structured agile development lifecycle with PM, Architect, Developer, and UX personas — is not fully replicated by any single existing PAI skill.
Gap Fill 1 PAI has no dedicated structured software development lifecycle workflow (requirements → PRD → architecture → implementation sprint cycles); BMAD addresses this gap directly, though the integration is workflow adoption rather than a missing infrastructure primitive.

Composite: 0.50

What Next

Landscape Position

Category: AI Agent Frameworks

In this category: VoltAgent--voltagent (AI Agent Engineering Platform, TypeScript, excellent 21/24, integrate)

Standing: BMAD-METHOD is the only development-methodology-focused agent framework in the vault; VoltAgent--voltagent is a general-purpose agent orchestration runtime — the two are complementary rather than competing, with BMAD supplying opinionated lifecycle workflow content and VoltAgent supplying programmable agent infrastructure.

Evidence Base

Density: 8/10 — Available: full README (8KB), complete package.json with all dependency and script data, stars/forks/open-issues metrics, release history with timestamps, last commit date, license metadata field, creation date. Missing: actual installed agent/workflow YAML content, CI workflow configuration files, contributor activity breakdown beyond aggregate star and fork counts.

Notes

The 4.6/24 pre-computation in the rolling summary is a placeholder estimate — careful probe-based scoring yields 19/24 (solid). The NOASSERTION license metadata is almost certainly a GitHub API artifact: both the README badge and package.json explicitly declare MIT. Repos at this adoption scale (48k stars, 5.6k forks in 13 months) warrant the quality signal reflected in the probe scores. The modular ecosystem design — a versioned core (BMM) extended by domain-specific modules (TEA for testing, BMGD for game dev, CIS for creative work) — is a meaningful architectural signal that distinguishes BMAD from single-purpose agent prompt packs and positions it as a genuine framework rather than a prompt library.