Skip to content

jeremylongshore/tons-of-skills-marketplace / plugins/ai-ml/jeremy-adk-software-engineer/skills/adk-engineer

adk-engineer

Verified

Execute software engineer specializing in creating production-ready ADK agents with best practices, code structure, testing, and deployment automation. Use when asked to "build ADK agent", "create agent code", or "engineer ADK application". Trigger with relevant phrases based on skill purpose.

FileKindSize
PRD.mdmarkdown2.8 KB
SKILL.mdskill md3.3 KB
references/ARD.mdmarkdown5.5 KB
references/errors.mdmarkdown5.9 KB
references/examples.mdmarkdown13.2 KB
references/implementation.mdmarkdown4.8 KB
scripts/setup-project.shexecutablescript2.3 KB
# PRD: ADK Engineer

**Version:** 2.1.0
**Author:** Jeremy Longshore
**Status:** Active
**Marketplace:** [tonsofskills.com](https://tonsofskills.com) by [Intent Solutions](https://intentsolutions.io)
**Portfolio:** [jeremylongshore.com](https://jeremylongshore.com)

---

## Problem Statement

ADK agent code that works in a prototype rarely survives production. Missing test coverage, inconsistent module boundaries, hardcoded tool configurations, and ad-hoc deployment scripts create agents that break on first contact with real traffic. Refactoring after the fact costs 3-5x more than building correctly upfront.

The ADK Engineer skill produces production-grade agent code with clean architecture, test coverage, and deployment automation from the start.

## Target Users

| User | Context | Primary Need |
|------|---------|-------------|
| ADK developers | Building new agents | Clean structure with tests from day one |
| Teams with existing agents | Refactoring for production | Module boundaries, test coverage, deployment plan |
| Platform engineers | Standardizing agent development | Repeatable patterns across teams |
| Solo developers | Shipping agent products | End-to-end from code to deployed, tested agent |

## Success Criteria

1. **Code structure**: Generated code follows single-responsibility modules with clear interfaces
2. **Test coverage**: Every tool function has unit tests; every agent has smoke prompt tests
3. **Incremental builds**: Each tool added independently with its own validation before wiring to agent
4. **Deployment readiness**: Generated code deploys without manual intervention to the target platform
5. **Operational safety**: Retries, timeouts, structured logging, and safe error messages built in

## Functional Requirements

1. Clarify agent requirements (goals, tools, constraints, deployment target)
2. Propose architecture (single vs multi-agent, orchestration, state management)
3. Scaffold project structure with proper module boundaries
4. Implement tools incrementally — one at a time, each with tests
5. Add operational guardrails (retries, backoff, timeouts, logging)
6. Validate locally (unit tests + smoke prompts)
7. Generate deployment plan with health checks

## Non-Functional Requirements

- Supports Python, Java, and Go runtimes
- Works with any ADK-compatible model provider
- Generates CI-compatible test commands
- Code follows language-specific conventions (PEP 8 for Python, etc.)

## Dependencies

- Target language runtime installed
- ADK SDK installed and configured
- Test runner available (pytest, JUnit, go test)
- GCP project access for deployment (optional)

## Out of Scope

- Model fine-tuning or training
- UI/frontend development
- Infrastructure provisioning (use jeremy-adk-terraform for that)
- Production monitoring setup (use jeremy-vertex-validator)
finishing-a-development-branchobra/superpowersUse when implementation is complete, all tests pass, and you need to decide how to integrate the workStars: 323kUpdated Claude CodeCodexVerifiedInstall
systematic-debuggingobra/superpowersUse when encountering any bug, test failure, or unexpected behavior, before proposing fixesStars: 323kUpdated Claude CodeCodexVerifiedInstall
test-driven-developmentobra/superpowersUse when implementing any feature or bugfix, before writing implementation codeStars: 323kUpdated Claude CodeCodexVerifiedInstall
skill-creatoranthropics/skillsCreate new skills, modify and improve existing skills, and measure skill performance. Use when users want to create a skill from scratch, edit, or optimize an existing skill, run evals to test a skill, benchmark skill performance with variance analysis, or optimize a skill's description for better triggering accuracy.Stars: 198kUpdated Claude CodeCodexVerifiedInstall