bugs-to-stories_skill

This skill converts bug reports into prd.json user stories for autonomous fixing, prioritizing critical issues and updating records for Ralph.

Shell

214

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 rohunj/claude-build-workflow --skill bugs-to-stories

SKILL.md5.5 KB

Overview

This skill converts bug reports produced by the test-and-break workflow into well-formed prd.json user stories so an autonomous fixer (Ralph) can address them. It reads the latest bug report, maps bug fields to FIX- story fields, prioritizes by severity, and updates or creates a prd.json ready for automated fixes. It ensures critical bugs are surfaced first and includes regression criteria for verification.

How this skill works

The skill reads the latest tasks/bug-report-*.md file, parses each bug entry, and transforms fields into a standardized story object (id, title, description, acceptanceCriteria, priority, passes, notes). It maps severity to priority ranges and either appends stories to an existing prd.json or generates a new bug-fix-only prd.json with an appropriate branchName. Finally it summarizes counts and priority distribution for user review.

When to use it

After running the test-and-break skill to convert its output into actionable work
When you need an ordered list of bug fix stories for an autonomous fixer (Ralph)
Before starting an automated fix run to ensure fixes are prioritized correctly
When you want regression criteria and typecheck gates added to every fix story
When creating a separate bugfix branch for an otherwise complete build

Best practices

Run on the latest bug report file (ls -t tasks/bug-report-*.md | head -1) to avoid stale data
Backup prd.json before modifying it (cp prd.json prd.json.backup) when appending
Prefer adding to existing prd.json if the original build is incomplete; create a bug-fix-only prd.json when the build is complete
Keep acceptanceCriteria explicit: technical fix steps, regression test referencing original steps, and typecheck pass
Ask the user whether to append or create a new prd.json if automation cannot decide

Example use cases

Convert multiple functional and UI bugs from automated testing into FIX- stories with priorities set
Create a new prd.json for a dedicated bugfix branch: ralph/bugfix-<date>
Append bug fix stories to an in-progress prd.json so Ralph continues the existing workflow
Generate acceptance criteria that include regression steps and typecheck requirements for each converted bug
Produce a summary report of how many fixes were added by priority bucket before launching the fixer

FAQ

Severity maps to a priority range: Critical 1-3, High 4-7, Medium 8-12, Low 13+. The skill assigns a concrete priority within that range and shifts by an offset when appending to an existing prd.json.

Will this overwrite my existing prd.json?

No — when appending, the skill recommends creating a backup (prd.json.backup) and appends stories after the current highest priority. It can also create a separate bug-fix-only prd.json if requested.

What acceptance criteria are included by default?

Each story includes explicit technical fix items (derived from the bug), a regression test that reproduces original steps, and a Typecheck passes criterion.