agentleFS
0xErwin10xErwin1/dbflux/skills/issue-creation/SKILL.md

issue-creation

Skill526 starsChanged 6 months agoApache-2.0View on GitHub

Create GitHub issues for DBFlux using the repository's real conventions. Trigger: When creating a GitHub issue, reporting a bug, requesting a feature, or documenting work that needs tracking.

7 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use chained-pr, work-unit-commits and 3 more.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • joeldevzSkill

    chained-pr

    Split large changes (>400 lines OR >5 slices) into a series of reviewable PRs. Two strategies — stacked-to-main and feature-branch-chain. Prevents the 100-conflict merge disaster.

    Also in 3 projects that use this

  • joeldevzSkill

    work-unit-commits

    Commit discipline. Each commit is a self-contained, reviewable work unit — code + tests + docs together. Stops the 190-commit PR disaster before it starts.

    Also in 3 projects that use this

  • alex-ilgayevSkill

    go-testing

    Handles all Golang testing tasks including running tests, writing new tests, and fixing test failures. Follows MCPSpy testing conventions with require for critical assertions and assert for non-critical ones.

    Also in 3 projects that use this

  • +2 more from joeldevz

The file

We split this skill into the rules it gives an agent

10 rules. 0 are common, 5 are distinctive, and 5 are very specific to this project.

  • 0

    Common

    Most of the 7 edited versions we read keep these.

  • 5

    Distinctive

    This author's own choices. Worth a look.

  • 5

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Check the label with gh issue view <N> --json labels.

    line 23
  2. Very specific:

    Use only labels from the taxonomy (*:bug / *:feature per area, driver, platform, data-model kind).

    line 31
  3. Distinctive:

    Search for duplicates before creating a new issue

    line 35
  4. Very specific:

    Keep the form's shape when filing through gh (the CLI shows no form): keep the field labels as ### <label> headings, and keep the Labels I …

    line 38
  5. Distinctive:

    Keep the issue grounded in observed repo behavior, failing commands, or a concrete user/problem statement

    line 39
  6. Very specific:

    Follow the Approval Gate above: a feature or behavior-change issue waits for status:approved.

    line 40
  7. Distinctive:

    Keep a feature proposal small enough to land in PRs of about 400 changed lines that each branch from main.

    line 41
  8. Very specific:

    Do not apply status:* labels yourself.

    line 42
  9. Distinctive:

    Do not invent milestones, assignees, labels, or project board fields unless the user explicitly asks

    line 43
  10. Distinctive:

    If key reproduction details or scope are missing, ask one concise follow-up question instead of guessing

    line 44

“Common” is measured: kept by at least 6 in 10 of the 7 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

When to Use

  • Creating a bug report for DBFlux
  • Proposing a feature or improvement
  • Opening a tracking issue before implementation work

Approval Gate

IssueBefore implementing
Feature or behavior change (feature_request.yml)Stop after filing. Start work only once a maintainer applies status:approved. status:needs-review means it is still being evaluated; status:rejected means do not build it.
Bug (bug_report.yml)No wait. Open the fix PR together with the issue and link it with Resolves #N.
Filed by the maintainerThe maintainer may skip the wait.

Check the label with gh issue view <N> --json labels. A comment that sounds positive is not approval.

Repository Facts

  • Issues are created from the forms in .github/ISSUE_TEMPLATE/: bug_report.yml for a defect, feature_request.yml for a capability. They are the only issue templates in the repo.
  • A form supplies its own title prefix ([bug] or [feature] ) and renders each field as a ### <label> section, ending with a Labels I plan to apply checklist.
  • CONTRIBUTING.md defines the contributor workflow (## Issues) and the label taxonomy (## Label Guide).
  • Labels live on the upstream repository, so a reporter without write access cannot apply them from a fork: fill in the label section the form provides, name the labels the issue needs, and ask a maintainer to apply them during triage.
  • Use only labels from the taxonomy (*:bug / *:feature per area, driver, platform, data-model kind). There is no area:* convention.

Critical Patterns

  • Search for duplicates before creating a new issue
  • Choose the form that matches the request: a defect goes to bug_report.yml, a new capability to feature_request.yml. Do not write a free-form body that skips the form's fields.
  • Fill every required field. bug_report.yml asks for the DBFlux version, operating system, OS detail, architecture, the affected driver, reproducible steps, expected and actual behavior, and logs; feature_request.yml asks for the problem, the proposed solution, alternatives considered, and the primary area.
  • Keep the form's shape when filing through gh (the CLI shows no form): keep the field labels as ### <label> headings, and keep the Labels I plan to apply checklist in the body.
  • Keep the issue grounded in observed repo behavior, failing commands, or a concrete user/problem statement
  • Follow the Approval Gate above: a feature or behavior-change issue waits for status:approved; a bug fix and an issue filed by the maintainer do not
  • Keep a feature proposal small enough to land in PRs of about 400 changed lines that each branch from main; if it cannot, say in the proposal how it splits (PRs are squash-merged, so avoid stacks)
  • Do not apply status:* labels yourself; maintainers set them during triage
  • Do not invent milestones, assignees, labels, or project board fields unless the user explicitly asks
  • If key reproduction details or scope are missing, ask one concise follow-up question instead of guessing

Composing the Body

A GitHub issue form turns each field into a ### <label> section and each checkbox list into a Markdown checklist, so the filed issue keeps that shape. Compose the body from the form's YAML rather than from memory — read .github/ISSUE_TEMPLATE/bug_report.yml or feature_request.yml and write the fields in the order it declares them.

A bug report has: DBFlux version, Operating system, OS / distro detail, Architecture, Affected driver (if any), Steps to reproduce, Expected behavior, Actual behavior, Logs / stack traces, Labels I plan to apply.

A feature request has: Problem, Proposed solution, Alternatives considered, Primary area, Labels I plan to apply.

Commands

# Search for possible duplicates
gh issue list --search "keyword"

# Read the fields of the form being used
sed -n '1,200p' .github/ISSUE_TEMPLATE/bug_report.yml

# Create the issue: the title carries the form's prefix, the body its fields
gh issue create --repo 0xErwin1/dbflux --title "[bug] <summary>" --body-file /tmp/issue.md

Resources

  • .github/ISSUE_TEMPLATE/bug_report.yml
  • .github/ISSUE_TEMPLATE/feature_request.yml
  • CONTRIBUTING.md
  • AGENTS.md
  • README.md
  • ARCHITECTURE.md
  • CODE_STYLE.md

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.