spec_generator

Generates structured Workable Spec JSON from GitHub issues to guide developer workers.

Updated Mar 22, 2026
One-click install
npx skills add https://github.com/bkrsna/gemini-cli --skill spec-generator-bkrsna
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: spec_generator
Source: https://github.com/bkrsna/gemini-cli/tree/main/tools/caretaker-agent/cloudrun/triage-worker/.gemini/skills/spec_generator
Command: npx skills add https://github.com/bkrsna/gemini-cli --skill spec-generator-bkrsna

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Turning a raw GitHub issue into an actionable implementation plan requires manual codebase exploration and careful specification writing, which is slow and error-prone for automated triage pipelines. ## Core Features & Use Cases - Structured Spec Generation: Extracts technical details from an issue and organizes them into a strict Workable Spec JSON schema with summary, implementation plan, and testing strategy. - Codebase-Grounded Planning: Requires verification that every file listed in files_to_modify actually exists in the repository, preventing hallucinated paths. - Pipeline-Safe Output: Enforces strict JSON formatting rules so downstream automated code generation does not break. - Use Case: In an automated issue-triage worker, feed a bug report and receive a complete spec listing the root cause, exact source files to change, ordered fix steps, and Vitest verification steps for a developer worker to execute. ## Quick Start Analyze this GitHub issue and generate a Workable Spec JSON with the implementation plan and testing strategy for the developer worker.

Frequently Asked Questions about spec_generator

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I convert a GitHub issue into an implementation plan automatically?

Provide the issue content and let the spec generator extract the problem, root cause, and context, then produce a Workable Spec JSON with files to modify, ordered fix steps, and a testing strategy. The output feeds directly into an automated developer worker.

What is a Workable Spec JSON schema?

It is a strict JSON object with issue_id, summary (problem, root cause, context), implementation_plan (files_to_modify, steps), and testing_strategy (test_file, expected_behavior, verification_steps, framework). Every field is required for the downstream pipeline.

Does the spec generator verify that file paths exist in the repository?

Yes. The instructions require codebase exploration first, and all files listed in files_to_modify must actually exist in the repository. Fabricated paths are explicitly prohibited.

Which testing frameworks does the spec generator support?

The testing_strategy framework field is restricted to Vitest or N/A. Test files are listed separately in testing_strategy.test_file and must never appear in files_to_modify.

Why does the generated spec break the downstream pipeline?

Failures occur when the output deviates from the schema, such as nesting objects inside the steps array or escaping single quotes with backslashes in JSON strings. The output must be standard, valid JSON with flat string arrays.