flow-next-qa — Claude Code skill: what it does, how to install
flow-next-qa drives a running app through scenarios pulled from the spec and logs P0/P1/P2 findings backed by captured evidence. Finishes with a ship verdict, yes or no.
MIT ★ 730 updated 2026-10-29
/plugin marketplace add gmickel/flow-next
/plugin install flow-next@flow-nextWhat flow-next-qa does
The skill turns the spec's acceptance criteria into test scenarios, maps R-IDs into a coverage table, and treats the boundaries section as a list of what not to test. It drives the app through flow-next-drive, and when a project has .flow/features/, it uses that for navigation between screens.
A ship verdict can only rest on evidence pulled from the live run: screenshots, console output, observed state. Marking something a pass after just reading the source code is explicitly off-limits; with no reachable app or no driver, the outcome is blocked rather than a pass. Reading source to explain a failure already caught on the live app is fine.
It's triggered by /flow-next:qa with a spec id, or runs as the optional pipeline.qa stage inside flow --auto (off by default). It sits alongside CI, staging and manual QA rather than replacing them, and it doesn't touch product code — only files findings and hands them off. That stage is worth turning on only where /flow-next:prime has confirmed the app is ready for QA: seeded data, a documented dev login, a surface that can actually be driven.
What it helps with
- run a finished feature through scenarios built from acceptance criteria
- log P0/P1/P2 findings backed by evidence from a live app
- get a ship-or-not verdict before make-pr
- check R-ID coverage against real test scenarios
- wire a first-pass QA check into an autonomous flow --auto pipeline
Example prompts for the agent
- Run /flow-next:qa against spec SPEC-042
- Use /flow-next:qa to QA the current feature and show me the verdict
- Turn on the pipeline.qa stage in flow --auto and check it with /flow-next:qa
How to install flow-next-qa
1Plugin marketplace
/plugin marketplace add gmickel/flow-next/plugin install flow-next@flow-nextThis installs the whole flow-next plugin with all its skills
2Skills CLI (Vercel)
npx skills add gmickel/flow-next --skill flow-next-qa -a claude-code3Manually: put the files in place
~/.claude/skills/flow-next-qa/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/flow-next-qa/
1Skills CLI (Vercel)
npx skills add gmickel/flow-next --skill flow-next-qa -a codex2Manually: put the files in place
.agents/skills/flow-next-qa/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add gmickel/flow-next --skill flow-next-qa -a cursor2Manually: put the files in place
.cursor/skills/flow-next-qa/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install gmickel/flow-next plugins/flow-next/skills/flow-next-qa2Manually: put the files in place
.github/skills/flow-next-qa/SKILL.md1Skills CLI (Vercel)
npx skills add gmickel/flow-next --skill flow-next-qa -a gemini-cli2Platform CLI
gemini skills install https://github.com/gmickel/flow-next --path plugins/flow-next/skills/flow-next-qaManually: put the files in place
.devin/skills/flow-next-qa/SKILL.mdThe legacy .windsurf/skills/ path still works
Claude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about flow-next-qa
How do I install flow-next-qa in Claude Code?
Run /plugin marketplace add gmickel/flow-next, then /plugin install flow-next@flow-next.
What does flow-next-qa do?
It drives a running app, checks scenarios derived from the spec, and logs P0/P1/P2 findings with captured evidence, ending in a ship verdict of yes or no.
Can flow-next-qa pass something just by reading the code?
No, that's explicitly forbidden: the verdict relies only on evidence from the running app, and without a reachable app or driver the run is blocked instead.
Source text
SKILL.md · MIT
---
name: flow-next-qa
description: Live-app real-user QA pass derived from the spec. Drives the running app via flow-next-drive, derives scenarios from the spec's AC / R-IDs / boundaries, files structured P0/P1/P2 findings with evidence, and ends with a YES/NO ship verdict receipt. Consumes `.flow/features/` navigation when present. Triggers on /flow-next:qa with a spec id. Runs user-invoked OR as the optional `pipeline.qa` stage of `flow --auto` (default off). Augments — never replaces — CI/staging/manual QA. FORBIDDEN from marking PASS by reading source — the verdict rests on captured evidence from the live app, never on agent narration.
user-invocable: false
allowed-tools: AskUserQuestion, Read, Bash, Grep, Glob, Write, Edit, Task
---
# /flow-next:qa — live-app real-user QA pass
Drive the running app the way an unforgiving real user would, with scenarios taken from the spec,
file P0/P1/P2 findings with captured evidence, and end with a YES/NO ship verdict written as a
`qa_verdict` receipt. The spec is the source of intent: acceptance criteria become scenarios,
R-IDs become the coverage table, boundaries say what not to test, decision context says what the
app should do.
QA is a cheap first live pass over the finished build. It does not replace CI, staging or manual
QA, and it does not fix product code: it files, surfaces and hands off. Findings are advisory; the
human reviewer and the land gate decide.
It runs when the user invokes it, or as the optional `pipeline.qa` stage (`off | on | auto`,
default `off`) that `/flow-next:flow` runs after all tasks are done and before make-pr. When
setting or reading that key, read [gate-selection.md](../flow-next-flow/references/gate-selection.md);
[`flowctl.md`](../../docs/flowctl.md) has the config row. Enable the stage only where
`/flow-next:prime` reports the app is QA-ready (seeded data, a documented dev login, a drivable
surface, readable runtime evidence); otherwise every run ends BLOCKED.
Read [working-rules.md](../../references/working-rules.md) first unless you already have this run; it holds for every step.
## The hard rule
**SHIP rests on evidence captured from the running app** (screenshots, console output, observed
state). A SHIP reached by reading the source or the diff, from narration, or from "the code looks
right" has broken this. With no reachable app or no driver, the outcome is BLOCKED, never a pass.
Reading source to explain a failure you already captured is fine.
## Preamble
flowctl is bundled, not on `PATH`. Define it once; later blocks here and in `workflow.md` use `$FLOWCTL`:
```bash
FLOWCTL="${DROID_PLUGIN_ROOT:-${CLAUDE_PLUGIN_ROOT}}/scripts/flowctl"
[ -x "$FLOWCTL" ] || FLOWCTL="<plugin-root>/scripts/flowctl" # <plugin-root> = the directory two levels above this skill's SKILL.md file (the harness gave you that file's absolute path when the skill loaded); substitute it literally
[ -x "$FLOWCTL" ] || FLOWCTL=".flow/bin/flowctl"
```
This skill runs inline on the host (not forked) so it can ask the user for undocumented facts.
## Arguments
The first non-flag token is the spec id. `--target <url>`, `--receipt <path>` and `--base <ref>`
take a value (as `--flag value` or `--flag=value`). `mode:autonomous` turns on autonomous mode.
```bash
SPEC_ID=""; PREV=""
# No positional parameters: the host rewrites them inside skill code blocks.
# Quote the pasted arguments once so a URL's `&` or `;` stays data.
RAW_ARGS="$ARGUMENTS"
for ARG in $RAW_ARGS; do
case "$PREV" in
--target) QA_TARGET_URL="$ARG"; PREV=""; continue ;;
--receipt) QA_RECEIPT_OVERRIDE="$ARG"; PREV=""; continue ;;
--base) QA_BASE_REF="$ARG"; PREV=""; continue ;;
esac
case "$ARG" in
--target|--receipt|--base) PREV="$ARG" ;;
--target=*) QA_TARGET_URL="${ARG#--target=}" ;;
--receipt=*) QA_RECEIPT_OVERRIDE="${ARG#--receipt=}" ;;
--base=*) QA_BASE_REF="${ARG#--base=}" ;;
mode:autonomous) QA_AUTONOMOUS=1 ;;
-*) echo "Unknown flag: $ARG (ignored)" >&2 ;;
*) [[ -z "$SPEC_ID" ]] && SPEC_ID="$ARG" ;;
esac
done
[[ -n "$PREV" ]] && echo "Flag $PREV given without a value (ignored)" >&2
[[ "${FLOW_AUTONOMOUS:-}" == "1" ]] && QA_AUTONOMOUS=1
NO_PROMPT=0
[[ "${QA_AUTONOMOUS:-}" == "1" || "${AUTONOMOUS:-}" == "1" ]] && NO_PROMPT=1
export QA_TARGET_URL QA_RECEIPT_OVERRIDE QA_BASE_REF QA_AUTONOMOUS NO_PROMPT
```
**Autonomous mode** (`NO_PROMPT=1`; `flow --auto` dispatches with `mode:autonomous`) asks
nothing. Each fact the interactive run would ask for is resolved from the spec, config or
environment; when it cannot be, the run writes a BLOCKED receipt and exits cleanly. The one
exception is an unresolvable spec id, which is an error (non-zero exit, message on stderr).
Now follow [workflow.md](workflow.md): discover, derive, prepare, execute, file, verdict.
© Gordon Mickel · MIT · gmickel/flow-next