learned-knowledge.mdx

title	Learned Knowledge
sidebarTitle	Learned Knowledge
description	Insights that transfer across users.
mode	wide

The Learned Knowledge Store captures reusable insights, patterns, and best practices that apply across users and sessions. Powered by semantic search, agents find and apply relevant knowledge automatically.

Aspect	Value
Scope	Configurable (global, user, or custom namespace)
Persistence	Long-term
Default mode	Agentic
Supported modes	Always, Agentic, Propose
Requires	Knowledge base with vector database

Prerequisites

Learned Knowledge requires a Knowledge base for semantic search:

from agno.knowledge import Knowledge
from agno.knowledge.embedder.openai import OpenAIEmbedder
from agno.vectordb.pgvector import PgVector, SearchType

knowledge = Knowledge(
    vector_db=PgVector(
        db_url="postgresql+psycopg://ai:ai@localhost:5532/ai",
        table_name="learned_knowledge",
        search_type=SearchType.hybrid,
        embedder=OpenAIEmbedder(id="text-embedding-3-small"),
    ),
)

Basic Usage

from agno.agent import Agent
from agno.db.postgres import PostgresDb
from agno.learn import LearningMachine
from agno.models.openai import OpenAIResponses

agent = Agent(
    model=OpenAIResponses(id="gpt-5.2"),
    db=PostgresDb(db_url="postgresql+psycopg://ai:ai@localhost:5532/ai"),
    learning=LearningMachine(
        knowledge=knowledge,
        learned_knowledge=True,
    ),
)

# User 1 saves an insight
agent.print_response(
    "Save this: When comparing cloud providers, always check egress costs first - "
    "they can be 10x different between providers.",
    user_id="alice@example.com",
)

# User 2 benefits from the insight
agent.print_response(
    "I'm choosing between AWS and GCP for our data platform. What should I consider?",
    user_id="bob@example.com",
)

Agentic Mode

The agent receives tools to manage knowledge explicitly.

from agno.learn import LearningMachine, LearningMode, LearnedKnowledgeConfig

agent = Agent(
    model=OpenAIResponses(id="gpt-5.2"),
    db=db,
    learning=LearningMachine(
        knowledge=knowledge,
        learned_knowledge=LearnedKnowledgeConfig(mode=LearningMode.AGENTIC),
    ),
)

Available tools: search_learnings, save_learning

The agent searches before answering questions and before saving (to avoid duplicates).

Propose Mode

The agent proposes learnings for user confirmation before saving.

from agno.learn import LearningMachine, LearningMode, LearnedKnowledgeConfig

agent = Agent(
    model=OpenAIResponses(id="gpt-5.2"),
    db=db,
    learning=LearningMachine(
        knowledge=knowledge,
        learned_knowledge=LearnedKnowledgeConfig(mode=LearningMode.PROPOSE),
    ),
)

agent.print_response(
    "That's a great insight about Docker networking. We should remember that.",
    user_id="alice@example.com",
)
# Agent proposes the learning, user confirms before it's saved

Always Mode

Learnings are extracted automatically after each response.

from agno.learn import LearningMachine, LearningMode, LearnedKnowledgeConfig

agent = Agent(
    model=OpenAIResponses(id="gpt-5.2"),
    db=db,
    learning=LearningMachine(
        knowledge=knowledge,
        learned_knowledge=LearnedKnowledgeConfig(mode=LearningMode.ALWAYS),
    ),
)

Tradeoff: extra LLM call per interaction, may save low-value insights.

Data Model

Field	Description
title	Short, searchable title
learning	The actual insight
context	When/where this applies
tags	Categories for organization
namespace	Sharing scope
user_id	Owner (if namespace="user")
created_at	When captured

What to Save

Good to save	Don't save
Non-obvious discoveries	Raw facts or data
Reusable patterns	User-specific preferences
Domain-specific insights	Common knowledge
Problem-solving approaches	Conversation summaries
Best practices	Temporary information

Good example:

"When comparing cloud providers, always check egress costs first - they vary dramatically (AWS: $0.09/GB, GCP: $0.12/GB, Cloudflare R2: free)."

Poor example:

"AWS has egress costs."

Accessing Learned Knowledge

lm = agent.learning_machine

# Search for relevant learnings
results = lm.learned_knowledge_store.search(query="cloud costs", limit=5)
for result in results:
    print(f"{result.title}: {result.learning}")

# Debug output
lm.learned_knowledge_store.print(query="cloud costs")

Context Injection

Relevant learnings are injected via semantic search:

<relevant_learnings>
**Cloud egress cost variations**
Context: When selecting cloud providers for data-intensive workloads
Insight: Always check egress costs first - they can be 10x different between providers.

**API rate limiting strategies**
Context: When designing APIs with high traffic
Insight: Use token bucket algorithm for rate limiting - it handles bursts better than fixed windows.
</relevant_learnings>

Namespaces

Control knowledge sharing:

from agno.learn import LearnedKnowledgeConfig

# Global: shared with all users (default)
learned_knowledge=LearnedKnowledgeConfig(namespace="global")

# User: private per user
learned_knowledge=LearnedKnowledgeConfig(namespace="user")

# Custom: team or domain-specific
learned_knowledge=LearnedKnowledgeConfig(namespace="engineering")

Combining with Other Stores

from agno.learn import LearningMachine

agent = Agent(
    model=OpenAIResponses(id="gpt-5.2"),
    db=db,
    learning=LearningMachine(
        knowledge=knowledge,
        user_profile=True,       # Who the user is
        user_memory=True,        # User's preferences
        learned_knowledge=True,  # Collective insights
    ),
)

Personalized responses drawing on collective knowledge.