Skip to main content

What Are Aspects?

Every fact in your CORE memory is automatically classified into one of 11 aspects. Think of aspects as filing cabinet drawers—when you need something specific, CORE knows exactly which drawer to open.

The 11 Aspects

1. Identity

What it captures: Who you are (slow-changing attributes) Example predicates: is, has, works at, lives in Example statement: “Manik works at Red Planet” When it surfaces: When Claude needs to understand your role, company, or background for context

2. Knowledge

What it captures: What you know and understand Example predicates: expert in, knows, understands, familiar with Example statement: “Expert in TypeScript and React” When it surfaces: When Claude is explaining technical concepts or choosing complexity level

3. Belief

What it captures: Why you think that way Example predicates: believes, values, thinks Example statement: “Values simplicity over complexity in code” When it surfaces: When Claude makes architectural decisions or design choices

4. Preference

What it captures: How you want things done Example predicates: prefers, likes, avoids, dislikes Example statement: “Prefers pnpm over npm” When it surfaces: During code reviews, setup, or when choosing tools/patterns

5. Action

What it captures: What you actually do (habitual behaviors) Example predicates: does, uses, practices Example statement: “Reviews PRs every morning before standup” When it surfaces: When suggesting workflows or understanding your routines

6. Goal

What it captures: What you want to achieve Example predicates: wants, aims for, plans to Example statement: “Launch MVP by Q2 2026” When it surfaces: When planning features, prioritizing work, or suggesting next steps

7. Directive

What it captures: Hard rules and automation triggers Example predicates: must, always, never, notify when Example statement: “Always run tests before creating a PR” When it surfaces: Always—directives are enforced rules that Claude must follow

8. Decision

What it captures: Past choices and their reasoning Example predicates: decided, chose, selected Example statement: “Chose Neo4j for graph storage because of relationship traversal performance” When it surfaces: When encountering similar technical choices or explaining architecture

9. Event

What it captures: Things that happened (with timestamps) Example predicates: attended, started, completed, happened on Example statement: “Started working on authentication refactor on Jan 15, 2026” When it surfaces: When you ask “what happened”, “when did we”, or need temporal context

10. Problem

What it captures: Challenges, blockers, issues you’ve faced Example predicates: has problem, blocked by, struggles with Example statement: “Hit rate limits with GitHub API when syncing 50+ repos” When it surfaces: When debugging similar issues or planning around known constraints

11. Relationship

What it captures: Connections between people, teams, projects Example predicates: works with, knows, reports to, collaborates with Example statement: “Works with Sarah on CORE project” When it surfaces: When understanding team dynamics or project collaborators

Why Aspects Matter

When you ask “What are my coding preferences?”, CORE doesn’t search through all your conversations. It:
  1. Classifies your query as an Aspect Query for Preference
  2. Filters statements where aspect = Preference
  3. Returns only preference facts: “Prefers TypeScript”, “Uses pnpm”, “Avoids class components”
Result: 5 precise facts in 300ms, instead of 50 maybe-relevant chunks in 2000ms.

How They’re Used Together

Different aspects surface at different times:
What Claude Is DoingAspects Surfaced
Writing codePreference, Directive, Knowledge
Code reviewPreference, Directive
Architectural decisionDecision, Belief, Problem
DebuggingProblem, Knowledge, Decision
Planning featuresGoal, Decision, Directive
Explaining conceptsKnowledge, Identity
This is how CORE provides right information, right time—not context dumping.