workflow

A set of rules for handling documentation requests through a structured process.

In plain words
What is it for?
Use it to inspect a project, identify the documents it needs, choose an appropriate level of detail, and check whether documented features still need to be built.
Why use it?
It prevents important context, existing documents, dependencies, or missing implementation work from being overlooked.

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/elirancv/cursormd/workflow
Clone the repo
git clone --depth 1 https://github.com/elirancv/CursorMD

Made for: Cursor.

Per session 1,493 This file is loaded in full into every session.
When invoked 1,493 The same file — it is already loaded in full.
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 $0.01493 $0.01493
Opus 5 $0.00746 $0.00746
Sonnet 5 $0.00299 $0.00299
Haiku 4.5 $0.00149 $0.00149

Measured yesterday against content hash 2465d9e0c678, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

workflow 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 yesterday.

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/workflow.mdc · 142 lines

How it starts

The opening of the file, as written. The whole thing — 142 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Documentation Workflow

Follow this structured approach for every documentation request.

Step 1: Clarify Context (Lightly)

Gather essential information:

  • Project name & domain
  • Stage (idea / in development / live)
  • Goal mode (Stability / MVP / Research / Dev Tooling / Balanced)
  • Tech stack (if relevant)
  • What document(s) they want now

Exception: If user says "just try something quickly", ask less and produce a useful draft fast.

Auto-Detection: Before proposing scope, automatically:

  1. Scan project for existing documentation files
  2. Check dependencies (see document-dependencies.mdc)
  3. Identify missing critical documents
  4. Detect blockers (missing dependencies)
  5. Auto-Implementation Check: If documentation exists, check if code/implementation is missing (see auto-implementation.mdc)
  6. Propose implementation: If docs exist but code missing, propose creating code according to specifications

Step 2: Propose Documentation Scope OR Implementation

If documentation exists but code is missing:

  • First: Show what documentation exists
  • Then: Show what code/implementation is missing according to docs
  • Propose: Creating code files according to specifications
  • Auto-pilot mode: Read docs and implement accordingly

If documentation is missing:

  • First: Show documentation status (what exists, what's missing, what blocks what)
  • Then: Suggest 1-5 docs with 1-line rationale each
  • Include: Dependency warnings if prerequisites are missing
  • For complex projects, offer a documentation roadmap first, then generate docs one-by-one
  • Show build order: The correct sequence (like puzzle pieces) based on dependencies

Step 3: Generate Document(s)

For Starter Pack Requests

When user asks for starter pack (e.g., "Create all documents needed for my idea"):

  1. Identify stage (idea / MVP / pre-launch / production)
  2. Show starter pack contents - List all documents that will be created
  3. Show build order - The sequence based on dependencies
  4. Ask for confirmation - "I'll create [X] documents for [stage]. Proceed?"
  5. Generate documents one by one in dependency order:
    • Show each document content
    • Ask for approval before creating
    • After each creation, update status and suggest next
    • Continue until all documents in pack are created

Read the full file on GitHub · 142 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. yesterday First seen · 142 lines · 1,493 tokens per session scan A 2465d9e0c678

Subscribe to this mod's changes

workflow is a cursor rule published in the GitHub repository elirancv/CursorMD (1 stars, last pushed 9mo ago), licensed MIT. It adds 1,493 tokens to every session, about $0.0075 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.