How do I install the Randroid Vibereview skill?

Run npx skillmds add randroids-dojo/randroid-vibereview in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is Randroid Vibereview safe to use?

This skill has not completed SkillMD's automated safety review yet. Capability flags: docs only. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Randroid Vibereview?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Randroid Vibereview free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

Who published Randroid Vibereview?

randroids-dojo (@randroids-dojo) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Randroid Vibereview by randroids-dojo

Name: Randroid Vibereview
Rating: 4.2 (5 reviews)
Author: randroids-dojo

Randroid Vibereview

Run VibeReview browser-game playtests, capture screenshot and browser-state evidence, and feed findings into Spiral-HTML review ledgers. Use when the user asks to playtest a browser game, capture VibeReview evidence, inspect an existing VibeReview session, or turn playtest findings into durable follow-ups.

randroids-dojo a2198d2 2 files · 2.8 KB Updated Sep 9, 2026

File contents

VibeReview

Use the VibeReview headless CLI to capture evidence from an actual browser-game playtest.

Preflight

Read the target repository's instructions and review ledgers.
Locate a headless vibereview binary. Prefer a project-relative build, then the known development checkout. Inspect unknown symlinks before executing them.
Reject GUI app-bundle executables and do not assume a bare vibereview command is the CLI.
Verify the selected binary with --version.
Open and play the game with the available browser-control capability before capturing evidence. Use an existing authenticated browser only when the session requires it.

Session workflow

Use the verified CLI path for all commands:

"<vibereview-cli>" start --project <path> [--title <title>]
"<vibereview-cli>" capture --project <path> --note <text> [--severity note|polish|issue|blocksRelease] [--rating 1...5] [--tags a,b,c]
"<vibereview-cli>" capture --project <path> --note-file <path-or->
"<vibereview-cli>" end --project <path>
"<vibereview-cli>" status [--project <path>]
"<vibereview-cli>" list

Start or inspect the intended session.
Capture only after observing the relevant state. Write concrete notes describing the action, observed result, and user impact.
Use blocksRelease only for evidence that genuinely blocks release.
End the session and inspect the written artifacts.
Check git status because VibeReview may add media and mutate review ledgers.

Spiral-HTML integration

Before classifying findings, read AGENTS.md, docs/PLAYTEST.html, docs/FOLLOWUPS.html, docs/OPEN_QUESTIONS.html, and docs/FUN_FACTOR_AUDIT.html when present.

Preserve append-only data-* entries.
Map severe evidence to a blocks-release follow-up unless the CLI already created one.
Put ambiguous product choices in OPEN_QUESTIONS.html with options and a recommended default.
Put experience-level gaps in FUN_FACTOR_AUDIT.html; do not mark implementation coverage complete merely because evidence was captured.
For legacy Markdown projects, preserve artifacts without silently converting the ledger format.

Finish with the session ID, captured evidence, ledger changes, follow-ups, and any unverified behavior.

---name: randroid-vibereviewdescription: Run VibeReview browser-game playtests, capture screenshot and browser-state evidence, and feed findings into Spiral-HTML review ledgers. Use when the user asks to playtest a browser game, capture VibeReview evidence, inspect an existing VibeReview session, or turn playtest findings into durable follow-ups.--- # VibeReview Use the VibeReview headless CLI to capture evidence from an actual browser-game playtest. ## Preflight 1. Read the target repository's instructions and review ledgers. 2. Locate a headless `vibereview` binary. Prefer a project-relative build, then the known development checkout. Inspect unknown symlinks before executing them. 3. Reject GUI app-bundle executables and do not assume a bare `vibereview` command is the CLI. 4. Verify the selected binary with `--version`. 5. Open and play the game with the available browser-control capability before capturing evidence. Use an existing authenticated browser only when the session requires it. ## Session workflow Use the verified CLI path for all commands: ```bash "<vibereview-cli>" start --project <path> [--title <title>] "<vibereview-cli>" capture --project <path> --note <text> [--severity note|polish|issue|blocksRelease] [--rating 1...5] [--tags a,b,c] "<vibereview-cli>" capture --project <path> --note-file <path-or-> "<vibereview-cli>" end --project <path> "<vibereview-cli>" status [--project <path>] "<vibereview-cli>" list ``` 1. Start or inspect the intended session. 2. Capture only after observing the relevant state. Write concrete notes describing the action, observed result, and user impact. 3. Use `blocksRelease` only for evidence that genuinely blocks release. 4. End the session and inspect the written artifacts. 5. Check `git status` because VibeReview may add media and mutate review ledgers. ## Spiral-HTML integration Before classifying findings, read `AGENTS.md`, `docs/PLAYTEST.html`, `docs/FOLLOWUPS.html`, `docs/OPEN_QUESTIONS.html`, and `docs/FUN_FACTOR_AUDIT.html` when present. - Preserve append-only `data-*` entries. - Map severe evidence to a `blocks-release` follow-up unless the CLI already created one. - Put ambiguous product choices in `OPEN_QUESTIONS.html` with options and a recommended default. - Put experience-level gaps in `FUN_FACTOR_AUDIT.html`; do not mark implementation coverage complete merely because evidence was captured. - For legacy Markdown projects, preserve artifacts without silently converting the ledger format. Finish with the session ID, captured evidence, ledger changes, follow-ups, and any unverified behavior.

Frequently asked questions