VeryGoodOpenSource / VeryGoodOpenSource/vgv-ai-flutter-plugin

feat: automated weekly skill check against Claude Code changelog

Open
#103 0 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

feature
Dominant language
Shell
Stars
162
Forks
23
Avg merge
2d 22h
Merged PRs (30d)
4

Description

Description

Add a bot-driven workflow that automatically checks this plugin's skills against the latest Claude Code changelog on a weekly cadence. When Claude Code ships changes that affect skill behavior (tool APIs, hook contracts, slash-command semantics, frontmatter conventions, MCP server contracts), the bot should surface which skills may be impacted so they can be reviewed and updated before they silently break.

Requirements

  • Scheduled weekly job (e.g. GitHub Actions cron) that fetches the latest Claude Code changelog/release notes.
  • Diff the changelog against the last-checked version and extract entries relevant to skills, agents, and MCP servers.
  • Map changed surfaces to potentially affected skills (e.g. bloc, testing, navigation, layered-architecture, create-project).
  • Output a report — open or update a tracking issue listing impacted skills and the relevant changelog entries.
  • No false-quiet: if parsing fails or the changelog format changes, the job flags it rather than passing silently.

Additional Context

Pairs with the evaluation tables/data work so impacted skills can be re-tested against updates automatically. See #104.

Contributor guide

Open the contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Start at the GitHub Actions workflow entry point and review how the repository organizes skills such as bloc, testing, navigation, layered-architecture, and create-project. Trace the Claude Code changelog/release notes input and the relationship to the evaluation tables/data work in #104. Done means a weekly report identifies affected skills, tracks the last checked version, and flags parsing failures instead of passing silently.

Written by the indexing model from the issue text.

Assessment

Tech stack
github-actions, shell
Domain
ci-cd, tooling
Issue type
Feature
Difficulty
5/5
Estimated time
Over a week
Activity status
Quiet
Clarity
Mostly clear
Newbie friendliness
42/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.