- Home
- Skills
- Gmickel
- Gmickel Claude Marketplace
- Flow Interview
flow-interview_skill
- Python
458
GitHub Stars
2
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 gmickel/gmickel-claude-marketplace --skill flow-interview- questions.md1.8 KB
- SKILL.md3.6 KB
Overview
This skill interviews you in-depth about a spec, bead, or feature idea to extract complete implementation details and produce a refined, actionable specification. It acts as a technical interviewer and spec refiner, guiding you through detailed, decision-focused questions and writing the updated bead or file back when complete.
How this skill works
Provide a bead ID, a file path, or leave the input empty to be prompted. The skill detects the input type, fetches the bead or reads the file, then runs a structured interview composed of many grouped questions using the AskUserQuestion tool. After the interview it writes a refined spec back into the bead or replaces the file, and summarizes questions asked, key decisions, and next steps.
When to use it
- You have an early feature idea and need to flesh out implementation details before building.
- You want to turn a vague spec, note, or bead into an actionable task with acceptance criteria.
- You need to refine requirements and capture technical decisions, edge cases, and dependencies.
- You plan to hand off work to engineers and want a complete spec to prevent rework.
Best practices
- Provide a bead ID or path to an existing spec to allow targeted, context-aware questioning.
- Be prepared for many focused questions; grouping of 2–4 related questions per step speeds the process.
- Answer decisively on core trade-offs (datastore, APIs, auth, scaling) to converge quickly.
- Use the summary and next-step suggestions to trigger planning or execution flows after the interview.
- Expect iteration: run the interview again after design changes or new constraints appear.
Example use cases
- Convert a one-paragraph feature idea into a task with technical approach, acceptance criteria, and edge cases.
- Refine an epic bead into clear subtasks with dependencies and implementation notes.
- Review and improve a draft SPEC.md file with missing details for authentication, error handling, and performance.
- Clarify API boundaries, data schemas, and migration steps before starting development.
FAQ
It accepts a bead ID (pattern like gno-42), a file path (e.g., docs/spec.md), or no input in which case it prompts you for a target.
How are interview questions delivered?
Questions are issued via the AskUserQuestion tool in grouped sets; the skill never prints individual questions as plain text.
What does it write back after the interview?
For beads it updates descriptions, acceptance criteria, and subtasks as needed. For files it rewrites the spec preserving structure and adding technical details, edge cases, and criteria.