bmad-agent-dev

Implements approved stories with test-first discipline as a senior software engineer persona.

Updated Mar 1, 2026
One-click install
npx skills add https://github.com/Hexalith/Hexalith.Parties --skill bmad-agent-dev-hexalith
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: bmad-agent-dev
Source: https://github.com/Hexalith/Hexalith.Parties/tree/main/.agents/skills/bmad-agent-dev
Command: npx skills add https://github.com/Hexalith/Hexalith.Parties --skill bmad-agent-dev-hexalith

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Teams following the BMad Method need a disciplined developer agent that executes approved stories with test-driven development, consistent persona behavior, and configurable activation steps instead of ad-hoc coding prompts. ## Core Features & Use Cases - Persona Activation: Adopts the Amelia senior software engineer identity with customizable role, communication style, principles, and persistent facts loaded from layered TOML configuration. - Menu-Driven Dispatch: Presents a numbered capability menu mapping codes to skills such as story implementation, E2E test generation, code review, sprint planning, and retrospectives. - Customization Merging: Resolves base, team, and user TOML overrides with structural merge rules so organizations can tailor behavior without editing the core skill. - Use Case: A developer says "hey Amelia, let's implement the next story" and the agent loads project config, greets the user in the configured language, and dispatches directly to the build workflow. ## Quick Start Ask the agent to implement the next approved story and it will activate the developer persona and dispatch the build workflow.

Frequently Asked Questions about bmad-agent-dev

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

FAQPage Schema
How do I customize the developer agent persona?

Edit the agent block in customize.toml or add team and user override TOML files under the project _bmad/custom directory. Scalars override, arrays append, and arrays of tables keyed by code replace matching entries.

How does the agent activation sequence work?

Activation resolves the agent customization block, runs prepend steps, adopts the persona, loads persistent facts and project config, greets the user, runs append steps, then presents or dispatches the menu.

What happens if the customization resolver script fails?

The skill falls back to manually reading the base customize.toml, team override, and user override files in order and applies the same structural merge rules itself.

Can I invoke a specific workflow without the menu?

Yes. If your initial message clearly maps to a menu item, such as asking to implement the next story, the agent skips the menu and dispatches that skill directly after greeting.

What are the limitations of the persona customization?

The agent name and title are hardcoded and cannot be changed through customization. Only role, identity, communication style, principles, facts, activation steps, and menu items are configurable.