knowledge-management-rule

Knowledge Management System

Rule for capturing, organizing, refining, and applying knowledge throughout the project lifecycle.

Learning refinement filters

• type: event pattern: "learning_create"
• type: event pattern: "learning_update"
• type: file_change pattern: ".cursor/learnings/*.md"
• type: command pattern: "knowledge"

Actionable insights filters

• type: command pattern: "insight"
• type: event pattern: "implementation_complete"
• type: event pattern: "task_complete"
• type: event pattern: "code_review"
• type: file_change pattern: "src/*"

actions:

SECTION 1: KNOWLEDGE CAPTURE

• type: react conditions:

  • pattern: "learn add|learning create" action: |

  Create a new learning entry

  I'll create a structured learning document with:

  • Unique ID (format: LEARN-YYYY-MM-DD-NN)
  • Title and short description
  • Detailed description of the learning (including context related to SOLID, TDD, SRP, Calisthenics, resilience if applicable)
  • Relevant code files (using Sources/, Tests/ paths), tasks (managed by development_workflow_system), specifications (managed by specification_management), and documents
  • Date of capture
  • Keywords for easy retrieval

  The learning will be stored in .cursor/learnings/ and indexed in .cursor/LEARNINGS.md

• type: react conditions:

  • pattern: "document add|document create|document detect" action: |

  Register a document in the knowledge base

  I'll register the document by:

  • Storing a copy in .cursor/docs/ if it's not already there
  • Creating a learning entry about this document
  • Adding the document to the documents index
  • Extracting key information from the document
  • Creating cross-references to related content including specifications (managed by specification_management)

  This ensures all project documentation is centrally tracked and discoverable.

• type: react event: "conversation_insight" action: |

  Automatically capture important insights from our conversations

  When I identify valuable insights in our conversation, I'll:

  • Create a learning entry with the insight
  • Generate a unique ID for the learning
  • Categorize and tag the insight appropriately
  • Add context and relevant connections to tasks (managed by development_workflow_system)
  • Index the insight for future reference

  This ensures important information isn't lost.

• type: react event: "file_change" conditions:

  • pattern: ".cursor/docs/.*" action: |

  Automatically process new documents added to docs directory

  When a new document is added to .cursor/docs/, I'll:

  • Extract key information from the document
  • Create a learning entry linked to this document
  • Update the documents index
  • Identify related content including specifications (managed by specification_management)
  • Generate appropriate metadata

  This ensures all documentation is properly integrated into the knowledge base.

SECTION 2: KNOWLEDGE ORGANIZATION & REFINEMENT

• type: react conditions:

  • pattern: "learn categorize|learning categorize" action: |

  Categorize and organize learnings

  I'll organize all learnings into meaningful categories:

  1. Analyze all learning content to identify topics and themes
  2. Categorize learnings into domains like:
     • Architecture, Performance, Security, DevOps
     • UX, API, Database, Testing
     • Frontend, Backend, Mobile, Tooling
     • Process, Bugs, Documentation
  3. Create category files in .cursor/learning_categories/
  4. Generate a categories index with links to all categorized learnings
  5. Identify uncategorized learnings for further review