Is the Flow Next Interview skill free?

Yes. The skill itself is free from gmickel/flow-next. SkillProof publishes the install command and an independent test verdict at no cost.

What is the Flow Next Interview SkillProof Score?

7.6/10 — installs cleanly 5/5, triggers reliably 5/5, output vs. baseline 5/10, docs & honesty 4/5.

How do I install Flow Next Interview?

Copy the install command from this page, run it in your terminal, and restart Claude Code. Skills live in ~/.claude/skills/ (global) or .claude/skills/ inside a project.

Can I use Flow Next Interview with Cursor, Copilot, Gemini CLI, Codex or other AI tools?

The SKILL.md format is native to Claude (Claude Code, Desktop, claude.ai). The instructions inside adapt to other assistants: Cursor rules, GitHub Copilot instructions, Windsurf rules, Custom GPTs, AGENTS.md for OpenAI Codex, and GEMINI.md for Google Gemini CLI — our conversion guides cover each, and the free converter on the tools page does the wrapping for you.

Flow Next Interview

Deep AskUserQuestion interview that refines a Flow spec, task, or spec file before you build

by gmickel · gmickel/flow-next

Works with setup

Install — copy the command Download source ↗

Test report

Verdict: Works with setup
Score: 7.6/10
Tested: Jul 21, 2026
Environment: Claude Code 2.x (agent harness)
Upstream re-checked: Aug 10, 2026 · 2b8cdb6

⚠ The author changed this skill after we tested it. The verdict below describes the version we ran on Jul 21, 2026; it's queued for a re-test.

Located the canonical SKILL.md at plugins/flow-next/skills/flow-next-interview/ (also mirrored under codex/). Frontmatter parses with name+description; it is user-invocable:false. Spot-checked 4 referenced files via raw fetch — flowctl launcher, questions-shared.md, references/write-back.md, templates/spec.md — all HTTP 200; no security smells (no curl|sh, base64, or exfiltration). I could NOT run the skill end-to-end: it needs the bundled flowctl CLI, a scaffolded .flow/ project (/flow-next:setup), and an interactive human for the 40-question AskUserQuestion loop — none available here, so outputMeasured=false and verdict=setup. I did hand-apply its question contract to one bare spec: versus my baseline flat list, the skill re-classified "what store backs the counters?" as codebase-answerable (resolve by grep, don't ask), rewrote the HTTP-status question to lead with "429 + Retry-After / Confidence: [high]" and per-option consequences, tagged the limit-value question [your-call] instead of inventing a recommendation, and added skip-handling that parks dismissed questions under ## Open Questions rather than writing them as decisions. That is a real methodology improvement, but it is my manual application, not the skill executing.

Scored on four weighted criteria — install, triggering, output vs. baseline, docs. How scoring works

Installs cleanly 5/5
Triggers reliably 5/5
Output vs. baseline 5/10
Docs & honesty 4/5

What Flow Next Interview does

Part of the flow-next plugin: conducts a thorough (40+ question) requirements interview about a Flow spec ID, task ID, or spec file, then writes the refined details back into the spec via the bundled flowctl CLI. Triggers on /flow-next:interview or when a user wants to flesh out a spec, refine requirements, or clarify a feature before planning. Scope-aware (business/technical/both) and doc-aware; questions lead with a recommendation plus a confidence tier and treat skipped questions as unresolved rather than silent decisions.

How to install Flow Next Interview

git clone --depth 1 https://github.com/gmickel/flow-next.git /tmp/flow-next-interview-src
mkdir -p ~/.claude/skills
cp -R /tmp/flow-next-interview-src/plugins/flow-next/skills/flow-next-interview ~/.claude/skills/flow-next-interview
# WARNING: this standalone copy is NOT functional on its own. The skill is
# user-invocable:false and belongs to the flow-next plugin (v3.0.0). Its body
# calls a BUNDLED CLI resolved via ${CLAUDE_PLUGIN_ROOT}/scripts/flowctl and
# requires a scaffolded .flow/ project. To actually run it:
#   1. Install the plugin from its marketplace:
#      /plugin marketplace add gmickel/flow-next   then   /plugin install flow-next
#   2. In your project run:  /flow-next:setup   (scaffolds .flow/ + bundles flowctl)
#   3. Invoke:  /flow-next:interview fn-1-add-oauth   (or a .md spec-file path)
# Requires: python3 (flowctl is a Python-backed bash launcher), jq, and an
# interactive human to answer the AskUserQuestion rounds.

Skills live in ~/.claude/skills/ (global) or .claude/skills/ (per-project). Restart Claude Code after installing.

Commands — how to trigger Flow Next Interview

/flow-next-interview Deep AskUserQuestion interview that refines a Flow spec, task, or spec file before you build

It also activates on plain-language prompts like these:

Interview me to flesh out the oauth spec
Refine requirements for fn-1-add-oauth before building
Clarify this feature spec through Q&A

Frequently asked questions

Is the Flow Next Interview skill free?: Yes. The skill itself is free from gmickel/flow-next. SkillProof publishes the install command and an independent test verdict at no cost.
Does Flow Next Interview work with Claude Code?: We tested it with Claude Code 2.x (agent harness) on Jul 21, 2026. Verdict: Works with setup. Located the canonical SKILL.md at plugins/flow-next/skills/flow-next-interview/ (also mirrored under codex/). Frontmatter parses with name+description; it is user-invocable:false. Spot-checked 4 referenced files via raw fetch — flowctl launcher, questions-shared.md, references/write-back.md, templates/spec.md — all HTTP 200; no security smells (no curl|sh, base64, or exfiltration). I could NOT run the skill end-to-end: it needs the bundled flowctl CLI, a scaffolded .flow/ project (/flow-next:setup), and an interactive human for the 40-question AskUserQuestion loop — none available here, so outputMeasured=false and verdict=setup. I did hand-apply its question contract to one bare spec: versus my baseline flat list, the skill re-classified "what store backs the counters?" as codebase-answerable (resolve by grep, don't ask), rewrote the HTTP-status question to lead with "429 + Retry-After / Confidence: [high]" and per-option consequences, tagged the limit-value question [your-call] instead of inventing a recommendation, and added skip-handling that parks dismissed questions under ## Open Questions rather than writing them as decisions. That is a real methodology improvement, but it is my manual application, not the skill executing.
What is the Flow Next Interview SkillProof Score?: 7.6/10 — installs cleanly 5/5, triggers reliably 5/5, output vs. baseline 5/10, docs & honesty 4/5.
How do I install Flow Next Interview?: Copy the install command from this page, run it in your terminal, and restart Claude Code. Skills live in ~/.claude/skills/ (global) or .claude/skills/ inside a project.
Can I use Flow Next Interview with Cursor, Copilot, Gemini CLI, Codex or other AI tools?: The SKILL.md format is native to Claude (Claude Code, Desktop, claude.ai). The instructions inside adapt to other assistants: Cursor rules, GitHub Copilot instructions, Windsurf rules, Custom GPTs, AGENTS.md for OpenAI Codex, and GEMINI.md for Google Gemini CLI — our conversion guides cover each, and the free converter on the tools page does the wrapping for you.