workflow-level4

A full development process for complex system work, covering setup, documentation, architecture decisions, design exploration, staged implementation, review, and archiving.

In plain words
What is it for?
Use it for large technical projects that need architecture planning, multiple implementation stages, creative or design work, ongoing assessment, and a lasting project record.
Why use it?
Complex changes can affect many parts of a system and are easy to lose track of. This process creates checkpoints and records the reasoning behind major decisions.

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/alkoleft/mcp-bsl-platform-context/workflow-level4
Clone the repo
git clone --depth 1 https://github.com/alkoleft/mcp-bsl-platform-context

Made for: Cursor.

Per session 0 Nothing until a file matches its globs; then the whole rule loads.
When invoked 3,729 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin 100% copy Near-identical to another mod 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 $0.00000 $0.03729
Opus 5 $0.00000 $0.01865
Sonnet 5 $0.00000 $0.00746
Haiku 4.5 $0.00000 $0.00373

Measured 3d ago against content hash 9b57eef061ff, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

workflow-level4 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 3d 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.

Origin

This is a copy

100% identical to workflow-level4 — 0 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.

.cursor/rules/isolation_rules/Level4/workflow-level4.mdc · 424 lines

How it starts

The opening of the file, as written. The whole thing — 424 lines — stays where its author put it; the contents beside it link to each section on GitHub.

COMPREHENSIVE WORKFLOW FOR LEVEL 4 TASKS

TL;DR: This document outlines a comprehensive workflow for Level 4 (Complex System) tasks, including 7 key phases with rigorous planning, mandatory creative phases, architectural design, phased implementation, and extensive documentation.

🔍 LEVEL 4 WORKFLOW OVERVIEW

graph LR
    Init["1. INITIALIZATION"] --> Doc["2. DOCUMENTATION<br>SETUP"]
    Doc --> Plan["3. ARCHITECTURAL<br>PLANNING"]
    Plan --> Create["4. CREATIVE<br>PHASES"]
    Create --> Impl["5. PHASED<br>IMPLEMENTATION"]
    Impl --> Reflect["6. REFLECTION"]
    Reflect --> Archive["7. ARCHIVING"]
    
    %% Document connections for each phase
    Init -.-> InitDocs["INITIALIZATION"]
    Doc -.-> DocDocs["DOCUMENTATION"]
    Plan -.-> PlanDocs["ARCHITECTURAL PLANNING"]
    Create -.-> CreateDocs["CREATIVE PHASES"]
    Impl -.-> ImplDocs["PHASED IMPLEMENTATION"]
    Reflect -.-> ReflectDocs["REFLECTION"]
    Archive -.-> ArchiveDocs["ARCHIVING"]

🔄 LEVEL TRANSITION HANDLING

graph TD
    L4["Level 4 Task"] --> Assess["Continuous<br>Assessment"]
    
    Assess --> Down["Downgrade to<br>Level 2/3"]
    Assess --> Split["Split into<br>Multiple Tasks"]
    
    Down --> L23Trigger["Triggers:<br>- Less complex<br>- Limited scope<br>- Few components"]
    
    Split --> MultiTrigger["Triggers:<br>- Too large<br>- Independent parts<br>- Parallel possible"]
    
    L23Trigger --> L23Switch["Switch to<br>Level 2/3 Workflow"]
    MultiTrigger --> CreateTasks["Create Multiple<br>Lower Level Tasks"]

Level 4 tasks involve complex systems that require comprehensive planning, rigorous design, systematic implementation, and thorough documentation. This workflow ensures all aspects are addressed with the appropriate level of detail, structure, and verification.

📋 WORKFLOW PHASES

Phase 1: INITIALIZATION

graph TD
    Start["Start Level 4 Task"] --> Platform{"Detect<br>Platform"}
    Platform --> FileCheck["Critical File<br>Verification"]
    FileCheck --> LoadStructure["Comprehensive Memory<br>Bank Structure Loading"]
    LoadStructure --> TaskCreation["Create Detailed<br>Task Framework"]
    TaskCreation --> Context["Establish Enterprise<br>Context"]
    Context --> Resources["Identify and Allocate<br>All Resources"]
    Resources --> SetupComplete["Initialization<br>Complete"]

Read the full file on GitHub · 424 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. 3d ago First seen · 424 lines · 3,729 tokens per session scan A 9b57eef061ff

Subscribe to this mod's changes

workflow-level4 is a cursor rule published in the GitHub repository alkoleft/mcp-bsl-platform-context (191 stars, last pushed 5mo ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 3,729 tokens. A static security scan graded it A with 0 findings. It is 100% identical to workflow-level4, differing in 0 lines, and is treated as a copy.