How I keep LLMs on a tight leash and stopped hand-creating 30 GitHub issues in the process
Luis Faria developed a Python CLI tool called gh-issue-creator to automate the creation of GitHub issues from planning documents, eliminating the need for manual, error-prone copy-pasting. The tool supports both JSON and markdown input formats and prevents duplicate issues by checking existing titles. It enforces structured, well-scoped issue creation, which is especially useful when working with AI agents to avoid scope hallucination.
- ▪Luis Faria created a 300-line Python CLI tool named gh-issue-creator to automate GitHub issue generation.
- ▪The tool accepts markdown or JSON input and skips duplicate issues by comparing titles with existing ones in the repository.
- ▪gh-issue-creator runs in dry-run mode by default and requires explicit confirmation to create issues via the --create flag.
- ▪The tool integrates with the GitHub CLI for authentication and API calls, requiring zero configuration on an authenticated machine.
- ▪Well-defined issue templates with acceptance criteria help prevent AI agents from hallucinating task scope during execution.
Opening excerpt (first ~120 words) tap to expand
try { if(localStorage) { let currentUser = localStorage.getItem('current_user'); if (currentUser) { currentUser = JSON.parse(currentUser); if (currentUser.id === 3112496) { document.getElementById('article-show-container').classList.add('current-user-is-article-author'); } } } } catch (e) { console.error(e); } Luis Faria Posted on May 17 How I keep LLMs on a tight leash and stopped hand-creating 30 GitHub issues in the process #python #github #productivity #opensource The way I keep LLMs on a tight leash is through structured issue breakdowns. In this post you'll see how I go from concept to issue breakdown to GitHub issues to code - and why that sequence makes it easier to run an orchestrator agent directing a focused executor agent without it hallucinating scope.
…
Excerpt limited to ~120 words for fair-use compliance. The full article is at DEV.to (Top).