bmad-technical-research

Conducts web-verified technical research and generates cited multi-section research reports.

Updated Jul 9, 2026
One-click install
npx skills add https://github.com/unbakedcosmos/thumbnailer --skill bmad-technical-research-unbakedcosmos
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: bmad-technical-research
Source: https://github.com/unbakedcosmos/thumbnailer/tree/main/.claude/skills/bmad-technical-research
Command: npx skills add https://github.com/unbakedcosmos/thumbnailer --skill bmad-technical-research-unbakedcosmos

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Producing a trustworthy technical research report requires gathering current web data, verifying claims across multiple sources, and structuring findings into a coherent document, which is slow and error-prone when done manually. ## Core Features & Use Cases - Guided Six-Step Workflow: Walks through scope confirmation, technology stack analysis, integration patterns, architectural patterns, implementation research, and final synthesis with user confirmation at each stage. - Web-Verified Citations: Requires live web search for every claim, applies multi-source validation and confidence levels, and cites URLs throughout the report. - Structured Report Generation: Produces a complete research document with executive summary, table of contents, recommendations, and source documentation saved to the planning artifacts folder. - Use Case: Ask for a technical research report on "GraphQL vs REST API architectures" and receive a step-by-step collaborative research session ending in a fully cited markdown report. ## Quick Start Ask the assistant to conduct a technical research report on a technology topic such as serverless deployment options for Node.js.

Frequently Asked Questions about bmad-technical-research

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

FAQPage Schema
How do I generate a technical research report with AI?▼

Start the skill and state your research topic, such as React vs Vue for large-scale applications. The workflow confirms scope, then runs six guided steps covering technology stack, integration patterns, architecture, implementation, and synthesis, producing a cited markdown report.

What topics can technical research workflows cover?▼

The workflow covers architecture analysis, implementation approaches, technology stacks, integration patterns, and performance considerations. Example topics include GraphQL vs REST, serverless deployment options, and framework comparisons.

Does technical research require web search access?▼

Yes, web search is a hard prerequisite and the workflow aborts if it is unavailable. Every claim must be verified against current public sources with URL citations and confidence levels rather than relying on training data alone.

Where is the technical research report saved?▼

The report is created in the planning artifacts folder under a research subfolder, named with the topic slug and date. Content is appended step by step, and the final synthesis adds an executive summary, table of contents, and recommendations.

Can I customize the technical research workflow behavior?▼

Yes, a customize.toml file supports activation steps prepended or appended to the workflow, persistent facts loaded from project files, and an on_complete hook. Team and user-level TOML overrides merge with the defaults.