composer-deep-research

composer-deep-research is a cursor rule for Cursor from madebyaris/cursor-composer-rules. It costs 28 tokens per session (756 once invoked), scanned A, original, MIT.

A research workflow for difficult investigations involving unfamiliar tools, comparisons, audits, or edge cases.

In plain words
What is it for?
It is for vendor comparisons, unfamiliar APIs, benchmarks, regulations, architecture studies, and security analyses based on cited primary sources.
Why use it?
It helps separate confirmed information from assumptions and deal explicitly with conflicts between sources.

Cursor rule for Cursor

Install

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.

agentmods
npx agentmods add rules/madebyaris/cursor-composer-rules/composer-deep-research
Clone the repo
git clone --depth 1 https://github.com/madebyaris/cursor-composer-rules

Made for: Cursor.

Wrote 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.

agentmods badge for composer-deep-research

README.md
[![agentmods](https://agentmods.dev/badge/rules/madebyaris/cursor-composer-rules/composer-deep-research.svg)](https://agentmods.dev/rules/madebyaris/cursor-composer-rules/composer-deep-research)
Your own site
<a href="https://agentmods.dev/rules/madebyaris/cursor-composer-rules/composer-deep-research"><img src="https://agentmods.dev/badge/rules/madebyaris/cursor-composer-rules/composer-deep-research.svg" alt="Measured on agentmods" height="20"></a>
Per session 28 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 756 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5 $0.00028 $0.00756
Opus 5 $0.00014 $0.00378
Sonnet 5 $0.00006 $0.00151
Haiku 4.5 $0.00003 $0.00076

Measured 5d ago against content hash 16a6162ab9e4, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

composer-deep-research 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 5d 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.

.cursor/rules/composer-deep-research.mdc · 66 lines

How it starts

The opening of the file, as written. The whole thing — 66 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Composer deep research

When the user asks for investigation, comparisons, unfamiliar domains, audits, benchmarks, regulation-style questions, vendor claims, historical timelines, architecture surveys, threat models grounded in citations, etc., shift into research mode. This is an edge tool, not the default build path.

Training data and prior familiarity are hypotheses until T1-anchored. Prefer primaries over recollection.

Expanded playbook: use the skill at .cursor/skills/deep-research/SKILL.md for ladders, contradiction logging, and long-form workflows.

Outcome shape

Produce answers in this order:

  1. Scoped conclusion framed to the user’s constraints (explicit assumptions).
  2. What primary sources establish versus what rests on weaker synthesis.
  3. Conflicts/disagreements between sources—with dates/versions/context.
  4. Action recommendation tied to assumptions and uncertainties.
  5. How to verify (commands, dashboards, reproducible benchmarks, authoritative doc sections).

Avoid confident absolutes unless directly supported—flag confidence per claim tier.

Visible research thinking

Do not expose private chain-of-thought or invented <thinking> blocks. Instead, externalize auditable reasoning artifacts:

  • Research plan: question, assumptions, source ladder, search angles, stop condition.
  • Claim ledger: important claim → source/evidence tier → counter-evidence checked → confidence.
  • Decision rationale: short “because” statements tied to evidence, not vibes.
  • Residual uncertainty: what would change the conclusion and how to test it.

For quick answers, compress these into 2–4 sentences. For deep research, keep them explicit.

Source ladder (default tiering)

Prefer, in descending trust for engineering facts unless the question is inherently opinion-heavy:

  1. Specifications, RFCs, official vendor docs keyed to product/version pages.
  2. Language/runtime/framework official guides and changelog entries—not random tutorials.
  3. Source repositories tagged releases, SECURITY advisories, CVE records with linked metadata.
  4. Peer-reviewed/academic/industry-neutral references when factual claims hinge on measurements.
  5. Secondary blogs/forums/stack answers—explicitly labelled as heuristic or anecdotal corroborating evidence.

Read the full file on GitHub · 66 lines

Changes

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.

  1. 5d ago First seen · 66 lines · 28 tokens per session scan A 16a6162ab9e4

Subscribe to this mod's changes

composer-deep-research is a cursor rule published in the GitHub repository madebyaris/cursor-composer-rules (3 stars, last pushed 17d ago), licensed MIT. It adds 28 tokens to every session and 756 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.