ss-agile-implementing-an-issue

To implement a single GitHub issue autonomously, you translate the issue body and acceptance criteria into code changes and commits on an already-checked-out branch. This enforces strict scope control by making conservative choices and reporting assumptions in a structured coordinator report.

Verifying code changes against acceptance criteria requires running relevant tests and checking each criterion individually during implementation. The process enforces a clean `git status --porcelain` state by committing generated outputs or updating `.gitignore`.

A subagent implementing GitHub issues can write code and create commits, but it must not pick issues, open pull requests, or merge. It operates within a pipeline to deliver merge-ready changes and returns a structured completion report instead.

Handling untracked files when finishing issue implementation requires enforcing a clean git tree with `git status --porcelain`. You must either commit generated outputs or update `.gitignore` to ensure no stray files remain before reporting completion.

Autonomous subagent coordination for issue implementation works by operating under a strict contract to never ask questions requiring human response. The subagent makes conservative decisions, implements the issue, verifies acceptance criteria, and outputs a parseable report with status, commits, and test results.

Automated issue implementation leaves a dirty git tree when generated outputs are not properly committed or ignored. Enforcing a clean `git status --porcelain` state requires committing build artifacts or updating `.gitignore` before returning the completion report.