Demand-First Review
Flags speculative or unused new API surface before reviewing implementation quality.
Test report
- Verdict
- Tested · Works
- Score
- Tested
- Jul 11, 2026
- Environment
- Claude Code 2.x (agent harness)
- Upstream re-checked
- Aug 10, 2026 · b2d27c3
Verbatim install (git clone + cp) worked with zero rescue: one valid SKILL.md, strict-YAML frontmatter, no referenced files, no API keys or external services. On an A/B review of a synthetic PR (new export + IPC channel + config flag, all with zero consumers) both arms independently caught the load-bearing finding — the new snapshot function exfiltrates internal-only flags across a documented security boundary — and reached identical remove/remove/defer decisions. The skill arm added its signature demand-normalization (all three surfaces express one premature future demand → consolidate) and an explicit step-7 gate that withholds implementation-quality review until surfaces survive; a modest prioritization edge over an already-strong baseline rather than a transformative one.
Scored on four weighted criteria — install, triggering, output vs. baseline, docs. How scoring works
- Installs cleanly 5/5
- Triggers reliably 4/5
- Output vs. baseline 6/10
- Docs & honesty 5/5
What Demand-First Review does
Use when reviewing a PR, API, IPC channel, endpoint, parameter, type, or config that adds new surface area — BEFORE commenting on implementation quality. Also use immediately when a review surfaces signals like "no consumers yet", "unused export", "speculative", "additive", "forward-compatible", or "for future use".
How to install Demand-First Review
git clone https://github.com/CherryHQ/cherry-studio
cd cherry-studio
mkdir -p ~/.claude/skills
cp -r .agents/skills/demand-first-review ~/.claude/skills/demand-first-review
Skills live in ~/.claude/skills/ (global) or .claude/skills/
(per-project). Restart Claude Code after installing.
Commands — how to trigger Demand-First Review
-
/demand-first-reviewFlags speculative or unused new API surface before reviewing implementation quality.
It also activates on plain-language prompts like these:
-
Review this new endpoint for unnecessary speculative fields -
Check if this new IPC channel has real consumers -
Is this config option actually needed right now
Frequently asked questions
- Is the Demand-First Review skill free?
- Yes. The skill itself is free from CherryHQ/cherry-studio. SkillProof publishes the install command and an independent test verdict at no cost.
- Does Demand-First Review work with Claude Code?
- We tested it with Claude Code 2.x (agent harness) on Jul 11, 2026. Verdict: Tested · Works. Verbatim install (git clone + cp) worked with zero rescue: one valid SKILL.md, strict-YAML frontmatter, no referenced files, no API keys or external services. On an A/B review of a synthetic PR (new export + IPC channel + config flag, all with zero consumers) both arms independently caught the load-bearing finding — the new snapshot function exfiltrates internal-only flags across a documented security boundary — and reached identical remove/remove/defer decisions. The skill arm added its signature demand-normalization (all three surfaces express one premature future demand → consolidate) and an explicit step-7 gate that withholds implementation-quality review until surfaces survive; a modest prioritization edge over an already-strong baseline rather than a transformative one.
- What is the Demand-First Review SkillProof Score?
- 8.0/10 — installs cleanly 5/5, triggers reliably 4/5, output vs. baseline 6/10, docs & honesty 5/5.
- How do I install Demand-First Review?
- 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 Demand-First Review 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.