- Home
- Skills
- Algiras
- Memory Palace
- Memory Palace Red Queen
memory-palace-red-queen_skill
- JavaScript
1
GitHub Stars
3
Bundled Files
2 months ago
Catalog Refreshed
4 months ago
First Indexed
Readme & install
Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.
Installation
Preview and clipboard use veilstrat where the catalogue uses aiagentskills.
npx veilstrat add skill algiras/memory-palace --skill memory-palace-red-queen- claude-plugin.json3.9 KB
- README.md8.6 KB
- SKILL.md8.2 KB
Overview
This skill transforms information into durable knowledge by combining the method of loci with continuous adversarial recall testing. It encodes concepts as vivid memory palaces and runs the Red Queen Protocol to detect and repair weak or hallucinated memories. Ideal for technical interview prep, certification study, and mastering complex subjects.
How this skill works
The skill encodes topics using SMASHIN SCOPE imagery into named memory palaces, creating strong anchors and sensory-rich scenes. Four agents (Examiner, Learner, Evaluator, Evolver) run adversarial recall cycles: generate hard questions, prompt blind recall, score accuracy against ground truth, and strengthen weak images. Continuous testing schedules and decay predictions keep memories accurate over time.
When to use it
- Preparing for technical interviews or timed Q&A sessions
- Studying for certifications or high-stakes exams
- Converting notes into long-term conceptual mastery
- Maintaining and auditing cross-topic knowledge over months
- Detecting and eliminating confidently held hallucinations
Best practices
- Always encode with SMASHIN SCOPE (sensory, absurd, interactive imagery) rather than rote notes
- Run Red Queen 'weak-spots' weekly to catch low-confidence items early
- Use interview mode under time pressure to simulate real exam/interview conditions
- Evolve failed recalls by redesigning anchors rather than re-reading material
- Keep a learning journal of gaps and remediation actions for each palace
Example use cases
- Create a "System Design" palace and store CAP theorem with a space-station theme, then run a weak-spots session to shore up gaps
- Encode algorithm invariants, run adversarial tests to expose edge-case misunderstandings before an interview
- Build a certification palace, schedule depth-first audits monthly, and use interview mode the week before the exam
- Cross-link related palaces and run cross-link strategy to ensure connections between concepts hold under recall pressure
FAQ
Weekly for active learning (weak-spots), daily in the final interview prep week, and monthly for full palace audits.
What if I keep failing the same memory?
Don’t re-read—have the Evolver create a stronger SMASHIN SCOPE image, change anchors, and retest with focused adversarial questions.
Can this detect hallucinations?
Yes. The Evaluator scores recall against ground truth and flags confident but incorrect recalls so you can correct false memories.