Composer Prompting
Shapes a coding task into a five-section Cursor/Composer delegation prompt
Test report
- Verdict
- Works with setup
- Score
- Tested
- Jul 29, 2026
- Environment
- Claude Code 2.x (agent harness)
- Upstream re-checked
- Jul 30, 2026 · 33e575a
⚠ This skill is no longer available upstream. Our re-check on Aug 10, 2026 couldn't find it any more (repo unreachable/deleted). The test below is what we measured on Jul 29, 2026 and we're leaving it up as a record — but there is nothing left to install, so we've removed the command.
Wrote a Cursor delegation prompt for a real repo twice, with and without the skill. Without it I ended the prompt at "run the test suite"; the skill's grounding step made me read package.json first, and executing that repo's real command chain showed a fresh clone dies on MODULE_NOT_FOUND until npm install runs, which is precisely the unverified "done" its anti-patterns file warns about. The five-section anatomy and the guardrails on lockfiles and pre-existing failures are concrete, and all three reference files exist and are substantive. The friction is that the frontmatter sets user-invocable: false, so this folder copied into ~/.claude/skills on its own can never be called; it is designed to be pulled in by the plugin's cursor-runner subagent, which means the real install is the whole plugin. The payoff also needs the Cursor CLI, which was not on PATH here, so the end-to-end delegation stayed unverified, and on any repo that already has an AGENTS.md the gain shrinks fast: this project's own AGENTS.md already spells out the same five sections.
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 7/10
- Docs & honesty 5/5
What Composer Prompting does
Guidance for turning a specified coding task into a prompt for the Cursor CLI, built around five mandatory sections (goal, repo context, acceptance criteria, files to touch, how to verify) plus a guardrails block. It also covers model escalation, chunking oversized plans, resume-versus-fresh routing, and a catalogue of prompt shapes that produce bad Composer runs. Intended to load inside the cursor-plugin-cc plugin before a /cursor:delegate handoff rather than being called directly by a user.
How to install Composer Prompting
Nothing to install: the source repository no longer has this skill. If the author brings it back, our daily re-check will pick it up and the command will reappear here.
Commands — how to trigger Composer Prompting
-
/composer-promptingShapes a coding task into a five-section Cursor/Composer delegation prompt
It also activates on plain-language prompts like these:
-
Delegate this refactor task to Cursor's Composer with a proper prompt -
Write the Composer prompt for step 3 of our implementation plan -
Turn this bug fix into a five-section prompt for /cursor:delegate
Frequently asked questions
- Is the Composer Prompting skill free?
- Yes. The skill itself is free from freema/cursor-plugin-cc. SkillProof publishes the install command and an independent test verdict at no cost.
- Does Composer Prompting work with Claude Code?
- We tested it with Claude Code 2.x (agent harness) on Jul 29, 2026. Verdict: Works with setup. Wrote a Cursor delegation prompt for a real repo twice, with and without the skill. Without it I ended the prompt at "run the test suite"; the skill's grounding step made me read package.json first, and executing that repo's real command chain showed a fresh clone dies on MODULE_NOT_FOUND until npm install runs, which is precisely the unverified "done" its anti-patterns file warns about. The five-section anatomy and the guardrails on lockfiles and pre-existing failures are concrete, and all three reference files exist and are substantive. The friction is that the frontmatter sets user-invocable: false, so this folder copied into ~/.claude/skills on its own can never be called; it is designed to be pulled in by the plugin's cursor-runner subagent, which means the real install is the whole plugin. The payoff also needs the Cursor CLI, which was not on PATH here, so the end-to-end delegation stayed unverified, and on any repo that already has an AGENTS.md the gain shrinks fast: this project's own AGENTS.md already spells out the same five sections.
- What is the Composer Prompting SkillProof Score?
- 8.4/10 — installs cleanly 5/5, triggers reliably 4/5, output vs. baseline 7/10, docs & honesty 5/5.
- How do I install Composer Prompting?
- 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 Composer Prompting 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.