CMMI-DEV Technical Solution

Overview

Draft CMMI-DEV Technical Solution (TS) deliverables: design documents and technical solution options for ML3.

Getting Started

Add your source materials to the project workspace, then activate the CMMI-DEV Technical Solution skill. The agent will guide you through the process step by step.

Example Conversation

You: We’re documenting our technical solution for the new search feature (build vs buy) and the chosen design for CMMI TS. We chose to extend our existing Elasticsearch setup. Can you draft the solution selection rationale and a short design document?

Agent: I’ll draft a Technical Solution document: (1) alternative solutions (e.g. build on existing Elasticsearch, buy SaaS search, build from scratch) and selection criteria; (2) selected solution and rationale; (3) design (architecture, interfaces, key components) and traceability to requirements. I’ll align with CMMI TS (solution selection, design, implementation).

You: Selection criteria were: time to market, TCO, and fit with current stack. We have a one-page architecture diagram in Confluence.

Agent: I’ve added the criteria and a short evaluation; selected solution “Extend Elasticsearch” with rationale. I’ve referenced the Confluence architecture and written a design summary (indexing pipeline, API, relevance tuning) in technical-solution-search.md. I can expand any section.

Sample Output Excerpt

Excerpt from a Technical Solution document produced with this skill.

Solution selection: Search capability

Alternatives: (1) Extend existing Elasticsearch; (2) SaaS search (e.g. Algolia); (3) Build custom (Lucene-based).

Criteria: Time to market (30%), TCO (40%), Fit with current stack (30%).

Selected: Extend existing Elasticsearch. Rationale: Already in use for logging; team has experience; fastest to market and lowest TCO; fits current stack. SaaS would reduce ops but increase cost and lock-in; custom build not justified for current scale.

Design summary

• Architecture: Application layer → search API → Elasticsearch cluster; indexing pipeline from main DB and CMS.
• Interfaces: Search API (REST): query, filters, pagination; indexing contract for content types.
• Key components: Indexer service, search API service, relevance/config (tuning, synonyms).