agent-memory-systems
Memory is the cornerstone of intelligent agents. Without it, every interaction starts from zero. This skill covers the architecture of agent memory: short-term (context window), long-term (vector stores), and the cognitive architectures that organize them. Key insight: Memory isn't just storage - it's retrieval. A million stored facts mean nothing if you can't find the right one. Chunking, embedding, and retrieval strategies determine whether your agent remembers or forgets. The field is fragm
Install with Tessl CLI
npx tessl i github:duclm1x1/Dive-Ai --skill agent-memory-systems28
Does it follow best practices?
If you maintain this skill, you can automatically optimize it using the tessl CLI to improve its score:
npx tessl skill review --optimize ./path/to/skillValidation for skill structure
Evaluation results
56%
Internal Knowledge Base Search System
Chunking and embedding pipeline design
Contextual chunking approach
94%
100%
Embedding model in metadata
25%
58%
Metadata filter before vector search
0%
0%
Chunk size evaluation
41%
16%
Retrieval quality focus
86%
80%
Embedding model consistency check
40%
30%
No fixed arbitrary chunking
100%
100%
Storage includes chunk metadata
50%
62%
Without context: $0.7383 · 5m 13s · 33 turns · 40 in / 10,740 out tokens
With context: $1.1672 · 7m · 49 turns · 166 in / 14,449 out tokens
100%
Personal Assistant Agent Memory Architecture
Memory type architecture and token budgeting
Multiple memory types used
100%
100%
Memory type rationale
100%
100%
Token budget allocation
100%
100%
Temporal scoring in retrieval
100%
100%
No single memory type for all data
100%
100%
Working memory separation
100%
100%
Context assembly documented
100%
100%
Memory decay or limits
100%
100%
Without context: $0.5724 · 3m 37s · 21 turns · 27 in / 11,763 out tokens
With context: $0.6617 · 3m 15s · 26 turns · 287 in / 11,332 out tokens
78%
15%Customer Support Agent Long-Term Memory
Memory storage and retrieval correctness
Conflict detection on write
90%
100%
Conflict resolution strategy
83%
100%
Memory decay or expiry policy
86%
100%
Retrieval-first design evidence
66%
66%
Temporal scoring in retrieval
0%
20%
Metadata filter in retrieval
40%
100%
No store-everything policy
90%
100%
Embedding model tracked
0%
0%
Without context: $0.7501 · 5m 29s · 26 turns · 31 in / 13,402 out tokens
With context: $0.7108 · 4m 57s · 33 turns · 70 in / 10,776 out tokens
- Evaluated
- Agent
- Claude Code
Table of Contents
Is this your skill?
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.