dev-kit-work_skill

This skill autonomously reads and implements tickets from .dev-kit/tickets, ensuring acceptance criteria are met and moved to completed when done.
  • Shell

0

GitHub Stars

1

Bundled Files

2 months ago

Catalog Refreshed

4 months ago

First Indexed

Readme & install

Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.

Installation

Preview and clipboard use veilstrat where the catalogue uses aiagentskills.

npx veilstrat add skill tom555my/dev-kit --skill dev-kit-work

  • SKILL.md4.8 KB

Overview

This skill implements existing tickets autonomously by reading ticket files in .dev-kit/tickets/, executing all acceptance criteria, and preparing the repository changes for review. It acts as a developer: parsing requirements, applying project standards, running tests, and producing a clear implementation summary. It stops before finalizing the ticket move and asks the user for confirmation to mark the ticket completed.

How this skill works

The skill loads a specified ticket file, extracts the user story, category, acceptance criteria, dependencies, and resources, then verifies prerequisites against other tickets and project documentation in .dev-kit/docs/. It implements each acceptance criterion by creating or modifying files, adding tests, and following repository patterns and type-safety rules. After testing and verification it summarizes changes, test results, and any follow-up work, then requests user approval to move the ticket to .dev-kit/tickets/completed/.

When to use it

  • You have a Feature, Bug, Enhancement, or Chore ticket in .dev-kit/tickets/ that should be implemented end-to-end.
  • You want an autonomous developer agent to write code, tests, and documentation following project patterns.
  • You need to enforce project standards (type safety, import aliases, styling patterns) during implementation.
  • You want the agent to surface discovered out-of-scope work and create follow-up tickets when necessary.
  • You need a reproducible implementation summary and a curated list of modified/created files.

Best practices

  • Only target non-Research tickets; Research tickets are redirected to the research skill.
  • Write clear, complete acceptance criteria in ticket files so the agent can implement without repeated clarifying questions.
  • Keep project docs up to date in .dev-kit/docs/ so the agent can verify standards and architecture.
  • Allow the agent to run tests and modify the repo but require user confirmation before moving tickets to completed.
  • When the agent discovers larger scope work, have it create follow-up tickets rather than silently expand the current task.

Example use cases

  • Implementing a UI feature described in a ticket with acceptance criteria and adding unit and integration tests.
  • Fixing a bug by following the ticket's reproduction steps, patching code, and adding regression tests.
  • Applying an enhancement that touches multiple modules while maintaining project import and styling conventions.
  • Completing a chore such as upgrading a dependency, updating configuration, and validating CI test runs.
  • Detecting and extracting out-of-scope refactors into new tickets with clear dependency links.

FAQ

Research tickets are redirected to the research skill immediately; only Feature, Bug, Enhancement, and Chore tickets proceed.

Will the agent move the ticket to completed automatically?

No. After implementation and verification it asks: "All acceptance criteria have been implemented. May I move this ticket to tickets/completed/?" and waits for your confirmation.

Built by
VeilStrat
AI signals for GTM teams
© 2026 VeilStrat. All rights reserved.All systems operational