Memory & Retention

Input: $ARGUMENTS

Step 1: Triage — What’s Worth Remembering?

Not everything deserves memory space. Classify the input:

MEMORY TRIAGE:
- [item 1] → [core concept / useful fact / reference / ephemeral] — because [reason]
- [item 2] → [core concept / useful fact / reference / ephemeral] — because [reason]
...

WORTH ENCODING: [list items that pass triage]
LOOK-UP INSTEAD: [list items better stored externally, with where to find them]

Step 2: Choose Encoding Strategy

For each item worth remembering, select the best encoding method:

• Elaboration: Connect to things you already know. “This is like X because…”
• Visual/Spatial: Create a mental image or place it in a memory palace
• Narrative: Embed in a story or sequence of events
• Chunking: Group related items into meaningful clusters
• Contrast: Define by what it is NOT, or compare to similar-but-different items
• Active recall: Turn it into a question you must answer from memory

ENCODING PLAN:
- [item]: Strategy = [method]. Encoding: [the specific elaboration/image/story/chunk]
...

RULE: The more connections, the stronger the memory. Isolated facts are fragile.

Step 3: Create Retrieval Cues

A memory is only useful if you can find it when you need it. For each item:

• When will I need this? (situation, trigger, context)
• What question will I be trying to answer? (retrieval prompt)
• What’s the minimum cue that would bring this back? (keyword, image, phrase)

RETRIEVAL CUES:
- [item]: Trigger = [when needed]. Cue = [minimal prompt]. Question = [what you'd ask]
...

Step 4: Connect to Existing Knowledge

New information sticks when it attaches to what you already know.

• What does this relate to that you already understand?
• Does it confirm, contradict, or extend existing knowledge?
• Where does it fit in your mental model of this domain?
• What does it make you reconsider?

CONNECTIONS:
- [item] connects to [existing knowledge] because [relationship]
- [item] changes my understanding of [topic] by [how]
...

KNOWLEDGE STRUCTURE:
[Brief map showing how new items connect to existing knowledge]

Step 5: Schedule Review

Memory decays predictably. Plan against the forgetting curve:

• Within 24 hours: First review — catches the steepest decay
• After 3 days: Second review — tests whether it stuck
• After 1 week: Third review — beginning of long-term consolidation
• After 1 month: Final review — if you still know it, it’s durable

REVIEW SCHEDULE:
- [item/group]: Review at [intervals based on importance]
- FORMAT: [active recall question / teach it to someone / apply it to a problem]

RULE: Passive re-reading is not review. You must actively retrieve.

Step 6: Prune Outdated Knowledge

Memory maintenance matters as much as memory formation.

• Is any existing knowledge now outdated because of what you’ve learned?
• Are you holding contradictory beliefs that need resolving?
• What can you now forget because this new knowledge supersedes it?

PRUNE LIST:
- [outdated item] — replaced by [new item]
- [contradiction] — resolve by [keeping X / updating Y]
...

SKIP: If nothing needs pruning, skip this step.

Step 7: Retention Summary

ENCODED:
- [item 1]: [strategy] — cue: [retrieval cue]
- [item 2]: [strategy] — cue: [retrieval cue]
...

EXTERNALIZED (look up, don't memorize):
- [item]: find at [location]

KEY CONNECTIONS:
- [most important link between new and existing knowledge]

NEXT REVIEW: [when and how]

Integration

Use with:

• /prcp -> Improve perception before deciding what to encode
• /mtcg -> Monitor whether you’re actually learning or just feeling familiar
• /lr -> Research a topic deeply, then use /memy to retain it
• /to -> Organize knowledge into a structured plan