Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add rules/madebyaris/cursor-composer-rules/composer-reasoninggit clone --depth 1 https://github.com/madebyaris/cursor-composer-rulesWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/rules/madebyaris/cursor-composer-rules/composer-reasoning)<a href="https://agentmods.dev/rules/madebyaris/cursor-composer-rules/composer-reasoning"><img src="https://agentmods.dev/badge/rules/madebyaris/cursor-composer-rules/composer-reasoning.svg" alt="Measured on agentmods" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00028 | $0.01019 |
| Opus 5 | $0.00014 | $0.00509 |
| Sonnet 5 | $0.00006 | $0.00204 |
| Haiku 4.5 | $0.00003 | $0.00102 |
Grade A, and why
composer-reasoning scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 4d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 78 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Composer reasoning
Use on high-stakes, architectural, or multi-option work — anywhere the cost of thinking wrong exceeds the cost of thinking twice. Basic intent, thread continuity, and the ask gate live in composer-core; this rule does not re-teach them and adds no gates to trivial work.
Companions: clarify-first (when to ask), composer-verification (proof after), composer-orchestration (plan mode).
Reason, then re-evaluate
First-draft thinking is a hypothesis. Run a second, adversarial pass before you commit:
- Draft the approach in one or two sentences.
- Critique it against the checklist below.
- Revise — or accept it and say why it survives.
- Decide, then act.
| Self-critique prompt | What it catches |
|---|---|
| What am I assuming that I haven't verified? | Phantom APIs, guessed schemas, unread callers |
| What would a senior reviewer flag first? | Blind spots you've normalized |
| What's the simplest thing that could work? | Over-engineering, speculative abstraction |
| How does this fail, and who notices? | Missing error/edge handling |
| If I'm wrong, how expensive is the undo? | One-way doors disguised as quick wins |
Make the loop visible and brief when it matters — "I considered X but chose Y because Z." Visible rationale is not a dump of private chain-of-thought; keep it short and decision-linked.
Calibrate, don't spiral. One quick pass for moderate change; a real second pass for a one-way door. If two passes don't converge, you're missing evidence — read the code or run the command instead of thinking harder.
Principal-level judgment
Weigh these before recommending:
- Tradeoffs, explicitly. Every choice spends something — latency, complexity, flexibility, time. Name what each option costs.
- Second-order effects. A cache adds invalidation. A new dependency adds upgrade burden. A clever abstraction taxes the next reader.
- Reversibility. Cheap-to-undo → speed. Hard-to-undo (migrations, public APIs, auth models, persisted formats) → slow second pass; often high confusion weight — ask if cues conflict (clarify-first).
- Cost of being wrong. A throwaway script and a billing path do not get the same scrutiny.
- The next engineer. Boring and obvious beats clever and surprising.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 4d ago First seen · 78 lines · 28 tokens per session scan A 1d63164ba7a0
composer-reasoning is a cursor rule published in the GitHub repository madebyaris/cursor-composer-rules (3 stars, last pushed 16d ago), licensed MIT. It adds 28 tokens to every session and 1,019 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other cursor rules, from other repositories
performance
Performance: profiling-first optimization, caching, bottleneck patterns.
security
Security: input validation, auth, data protection.
prisma
Prisma: schema modeling, queries, migrations.
supabase
Supabase: RLS, edge functions, realtime.
documentation
Documentation: code docs, READMEs, ADRs, maintenance.
git-workflow
Git workflow: commits, branches, PRs, history management.