AI SummaryCursor Rules for coordinating multi-agent workflows in book production, including em dash replacement, deep research, and integrated publishing processes. Developers managing complex document production pipelines with multiple specialized agents benefit from this structured ruleset.
Install
Copy this and paste it into Claude Code, Cursor, or any AI assistant:
I want to add the "wellspring_directory — Cursor Rules" prompt rules to my project. Repository: https://github.com/JeromyJSmith/wellspring_directory Please read the repo to find the rules/prompt file, then: 1. Download it to the correct location (.cursorrules, .windsurfrules, .github/prompts/, or project root — based on the file type) 2. If there's an existing rules file, merge the new rules in rather than overwriting 3. Confirm what was added
Description
Cursor Rules for wellspring_directory
Tool Usage Matrix
• File Operations: filesystem MCP (all agents) • Web Research: brave-search, firecrawl, puppeteer (DeepResearchAgent) • Data Storage: sqlite MCP (DataLayerAgent, TaskTracker) • External Integration: notion MCP (DocBot, TaskTracker) • Complex Reasoning: sequential-thinking MCP (PlanningArchitect) • Time Management: time MCP (TaskTracker, scheduling)
Core Workflow Philosophy
Reference: global_rules.mdc and agent_roles.mdc
1. Em Dash Replacement Workflow
Orchestrating Agent: PlanningArchitect Executing Agent: TypographyAgent Supporting Agents: DataLayerAgent (Roo), RuleKeeper Process: • PlanningArchitect receives em dash replacement request • TypographyAgent validates input files in em_dash_replacement/input/ • DataLayerAgent (Roo) defines data access patterns for file operations • TypographyAgent applies rules from em_dash_rules.md • Output processed files to em_dash_replacement/output/ • RuleKeeper validates compliance with typography_rules.md • DocBot updates logs and documentation MCP Tools Used: filesystem, sqlite (for tracking)
2. Deep Research Workflow
Orchestrating Agent: PlanningArchitect Executing Agent: DeepResearchAgent Supporting Agents: DataLayerAgent (Roo), TaskTracker Process: • PlanningArchitect breaks down research requirements • DeepResearchAgent follows research_rules.md • Execute web research using brave-search and firecrawl MCP tools • DataLayerAgent (Roo) manages citation database and extracted statements • Store results in deep_research_agent/ subdirectories: • citations/ - Source citations and references • extracted_statements/ - Key findings and quotes • reports/ - Comprehensive research reports • logs/ - Process documentation • TaskTracker updates progress and deliverable status MCP Tools Used: brave-search, firecrawl, puppeteer, sqlite, filesystem
Discussion
Health Signals
My Fox Den
Community Rating
Sign in to rate this booster
Works With
Any AI assistant that accepts custom rules or system prompts