Is the Flow Next Plan 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 Plan SkillProof Score?

8.8/10 — installs cleanly 5/5, triggers reliably 5/5, output vs. baseline 7/10, docs & honesty 5/5.

How do I install Flow Next Plan?

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 Plan 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 Plan

Turns feature requests or Flow IDs into structured .flow specs with sized tasks, no code

by gmickel · gmickel/flow-next

Works with setup

Install — copy the command Download source ↗

Test report

Verdict: Works with setup
Score: 8.8/10
Tested: Jul 21, 2026
Environment: Claude Code 2.x (agent harness)
Upstream re-checked: Aug 10, 2026 · ed27f9e

⚠ 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.

Cloned the repo and ran the bundled flowctl (Python) in a temp HOME: `init` scaffolded .flow/, `spec create` + `spec set-plan` + `task create` produced a real spec and task, and `validate --spec` returned valid:true, task_count:1. Built two artifacts for "rate limit the login endpoint": a baseline plan (15-line copy-paste JS rateLimit implementation, one task lumping 9 concerns, acceptance = "it works and blocks too many logins") vs. the skill-followed spec (signature only `function rateLimit(req,res,next):void`, explicit Scope IN/OUT, five numbered testable R1-R5 acceptance criteria, a single M-sized task, a surfaced open question) — the skill version validated clean via flowctl. Could NOT run the mandatory parallel scout fan-out (repo-scout/docs-scout are plugin subagents), so file:line reuse refs stayed as placeholders; that plus the plugin-runtime dependency (flowctl only resolves after installing the whole plugin and running /flow-next:setup) is the setup friction. All referenced files (steps.md, examples.md, references/html-artifacts.md, scripts/flowctl) returned HTTP 200; no security smells — no curl|sh, base64 blobs, or exfiltration.

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 7/10
Docs & honesty 5/5

What Flow Next Plan does

A planning skill for the flow-next plugin that converts a feature request or an existing Flow ID into a structured spec plus right-sized tasks stored in .flow/ via a bundled flowctl CLI, enforcing testable R-ID acceptance criteria and a no-implementation-code rule. Triggers on the /flow-next:plan slash command with a text description or a Flow ID (fn-1-add-oauth, fn-1-add-oauth.2). It plans only; implementation happens later in /flow-next:work.

How to install Flow Next Plan

git clone --depth 1 https://github.com/gmickel/flow-next.git /tmp/flow-next-plan-src
mkdir -p ~/.claude/skills
cp -R /tmp/flow-next-plan-src/plugins/flow-next/skills/flow-next-plan ~/.claude/skills/flow-next-plan
# IMPORTANT: this skill is a COMPONENT of the flow-next plugin and is NOT functional as a
# lone skill copy. It calls a bundled flowctl at $CLAUDE_PLUGIN_ROOT/scripts/flowctl (unset
# for a standalone skill) then falls back to .flow/bin/flowctl, which only exists AFTER setup.
# Proper install is via the plugin marketplace, then run setup to scaffold .flow/ and flowctl:
#   /plugin marketplace add gmickel/flow-next
#   /plugin install flow-next
#   /flow-next:setup          # creates .flow/ and .flow/bin/flowctl
# Invoke (skill is user-invocable:false — driven by the command): /flow-next:plan <feature or fn-ID>
# Deps: python3 (flowctl runtime), jq. flowctl is zero-dependency pure-stdlib Python.

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

Commands — how to trigger Flow Next Plan

/flow-next-plan Turns feature requests or Flow IDs into structured .flow specs with sized tasks, no code

It also activates on plain-language prompts like these:

Plan the implementation for the fn-1-add-oauth spec
Create a build plan from this feature request
Design the implementation steps for this Flow spec

Frequently asked questions

Is the Flow Next Plan 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 Plan work with Claude Code?: We tested it with Claude Code 2.x (agent harness) on Jul 21, 2026. Verdict: Works with setup. Cloned the repo and ran the bundled flowctl (Python) in a temp HOME: `init` scaffolded .flow/, `spec create` + `spec set-plan` + `task create` produced a real spec and task, and `validate --spec` returned valid:true, task_count:1. Built two artifacts for "rate limit the login endpoint": a baseline plan (15-line copy-paste JS rateLimit implementation, one task lumping 9 concerns, acceptance = "it works and blocks too many logins") vs. the skill-followed spec (signature only `function rateLimit(req,res,next):void`, explicit Scope IN/OUT, five numbered testable R1-R5 acceptance criteria, a single M-sized task, a surfaced open question) — the skill version validated clean via flowctl. Could NOT run the mandatory parallel scout fan-out (repo-scout/docs-scout are plugin subagents), so file:line reuse refs stayed as placeholders; that plus the plugin-runtime dependency (flowctl only resolves after installing the whole plugin and running /flow-next:setup) is the setup friction. All referenced files (steps.md, examples.md, references/html-artifacts.md, scripts/flowctl) returned HTTP 200; no security smells — no curl|sh, base64 blobs, or exfiltration.
What is the Flow Next Plan SkillProof Score?: 8.8/10 — installs cleanly 5/5, triggers reliably 5/5, output vs. baseline 7/10, docs & honesty 5/5.
How do I install Flow Next Plan?: 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 Plan 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.