develop

Implements stories and technical tasks with status validation, risk gating, and test-driven checkpoints.

2|Updated Feb 14, 2026
One-click install
npx skills add https://github.com/Gamaroff/agent-skills --skill develop-gamaroff
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: develop
Source: https://github.com/Gamaroff/agent-skills/tree/main/skills/develop
Command: npx skills add https://github.com/Gamaroff/agent-skills --skill develop-gamaroff

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve? Turning a written story or task document into working, tested code requires disciplined sequencing: validating the document is ready, checking for conflicting existing implementations, writing tests, and recording what was built. This Skill enforces that workflow so implementation stays aligned with requirements and fully documented. ## Core Features & Use Cases - Document-driven implementation: Detects story, task, or epic files by naming pattern, validates status (Draft, Ready for Development, In Progress), and blocks development of epics by routing to story or task creation first. - Risk and alignment gates: Flags high-risk stories (auth, payments, crypto) and offers to run QA planning first; analyzes existing code against the document and resolves misalignment with the user before writing code. - Test-gated task completion: Runs lint and test checkpoints per task, captures failures to log files with triage, applies a three-strikes escalation rule, and only checks off tasks when all tests pass. - Use Case: Run it on story.178.8.example-feature.md and the agent validates the story, checks risk level, aligns with any existing code, implements each task with co-located tests, and maintains a complete Dev Agent Record with file lists and change logs. ## Quick Start Ask the agent to develop the story file at stories/story.178.8.example-feature/story.178.8.example-feature.md and follow its guided implementation workflow.

Frequently Asked Questions about develop

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

FAQPage Schema
How do I implement a user story with an AI coding agent?▼

Invoke the develop workflow with the story file path, such as story.178.8.example-feature.md. The agent validates the story status, checks risk level, analyzes existing code alignment, then implements each task with tests before marking checkboxes complete.

How to resume a partially completed development task?▼

Re-invoke the workflow on the same story or task file. The agent counts checked versus unchecked task checkboxes, logs the resumption point, and continues from the first unchecked item without redoing completed work.

Can I develop an epic file directly?▼

No, epic files cannot be developed directly. When an epic is detected, the workflow halts and offers to create a story or task from it first, since only stories and tasks are implementable units.

What happens when existing code conflicts with the story requirements?▼

The alignment analysis compares existing implementation against the document and reports it as aligned, partially aligned, or misaligned. You then choose to align code to the document, align the document to code, take a hybrid approach, or decide manually.

Why does development halt after repeated test failures?▼

A three-strikes escalation rule applies: after the third consecutive failure, the workflow halts and documents the triage summaries, fix strategies tried, and remaining root causes, then asks you for guidance instead of spinning indefinitely.

When should QA planning run before development?▼

When a story frontmatter sets risk_level to high, covering areas like auth, payments, or cryptographic operations, the workflow recommends running QA planning first to catch architecture and security issues before code is written.