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 s...
SkillScore96Maintained
Automatic
◆AI Engineer pack
Last verified 8/18/2026
What it changes
Passed 22 of 23 test cases with the skill, and 20 of 23 without it.
Without skill87%
With skill96%
Checks it passed that the model alone missed
includes conflict detection when writing new entries to memory
incorporates temporal scoring to account for memory recency and decay
requires tracking the embedding model identifier in document metadata
and 2 more in the full report
3 fixed19 passed either way0 still failing1 broke
Pass-rate lift
+9 pp
Tokens
+10%
Agent turns
0%
Response time
-6%
Model gemini-3.6-flashMeasured 8/18/2026v122 of 23 cases passed with skill
Tokens, turns and time describe resource use, not a dollar price.
See it in action
One of 3 test cases the skill fixed. Same model, same prompt, with and without the skill.
1 of 3
Prompt
I am building a multi-turn support agent for an enterprise platform and need a complete memory system architecture design. Please provide a formal technical breakdown covering how to structure working memory, episodic history, and long-term knowledge storage. Include specific guidance on managing context window constraints across memory tiers and optimizing retrieval accuracy for user queries.
What was checked, without and with the skill
Checked
Without skill
With skill
recommends setting explicit token budgets across working, episodic, and long-term memory tiers
Passed
Passed
incorporates temporal scoring to account for memory recency and decay
Missed
Passed
includes conflict detection when writing new entries to memory
Missed
Passed
case-01 · gemini-3.6-flash · 8/18/2026 · answers are capped and redacted
What's inside
The instructions your model reads when the skill loads.
SKILL.md64 linessingle file
Agent Memory Systems
You are a cognitive architect who understands that memory makes agents intelligent.
You've built memory systems for agents handling millions of interactions. You know
that the hard part isn't storing - it's retrieving the right memory at the right time.
Your core insight: Memory failures look like intelligence failures. When an agent
"forgets" or gives inconsistent answers, it's almost always a retrieval problem,
not a storage problem. You obsess over chunking strategies, embedding quality,
and
Capabilities
agent-memory
long-term-memory
short-term-memory
working-memory
episodic-memory
semantic-memory
procedural-memory
memory-retrieval
memory-formation
memory-decay
Patterns
Memory Type Architecture
Choosing the right memory type for different information
Vector Store Selection Pattern
Choosing the right vector database for your use case
Create/update .agent/constitution.md. Use when commands/boundaries/constraints must be confirmed before baseline or code changes. Draft v0 from repo evidence, then interview user.