Institutional learning

Share what every review learns so the next one starts from the team’s accumulated judgment.

A research team that does not learn re-derives the same conclusions and repeats the same mistakes. The value compounds when an insight from one review is available to every agent on the next. Agno’s LearningMachine with a shared store and a global namespace puts every agent’s learning in one place the whole team reads from.

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

# team_learnings is a Knowledge instance backed by a vector database, on its own
# table so learnings stay out of the research library's search results.
from agents.settings import team_learnings

db = PostgresDb(db_url="postgresql+psycopg://ai:ai@localhost:5532/ai")


def institutional_learning() -> LearningMachine:
    return LearningMachine(
        knowledge=team_learnings,
        learned_knowledge=LearnedKnowledgeConfig(
            mode=LearningMode.AGENTIC,
            namespace="global",
        ),
    )


learning_instructions = (
    "Use save_learning to store reusable insights. "
    "Use search_learnings to find prior knowledge before forming a view."
)

analyst = Agent(
    name="Financial Analyst",
    model=OpenAIResponses(id="gpt-5.5"),
    db=db,
    instructions=learning_instructions,
    learning=institutional_learning(),
)

risk_officer = Agent(
    name="Risk Officer",
    model=OpenAIResponses(id="gpt-5.5"),
    db=db,
    instructions=learning_instructions,
    learning=institutional_learning(),
)

# The risk officer captures an insight:
risk_officer.print_response(
    "Save this insight: analyst estimates lag this sector by about a quarter."
)

# Later, a different agent reads it back from the shared store:
analyst.learning_machine.learned_knowledge_store.print(query="estimate lag")

The learning is available to every agent and team sharing the store. In Agentic mode, retrieve it with search_learnings. The Knowledge instance is required. Without it, saves and searches log a warning and do nothing.

Per-agent vs institutional

	User memory	Institutional learning
Scope	One user across agents sharing the database	Every agent and team that shares the store
Namespace	Scoped by user_id	"global"
Effect	The user’s preferences and facts follow them	The committee remembers

For deep research, institutional is the point. The team’s judgment should outlive any single review.

What to capture

Worth a learning	Not worth a learning
”Analyst estimates lag this sector by a quarter”	A one-off number from a single query
”This data source double-counts renewals”	A restatement of the mandate
A correction to a conclusion that was wrong	A summary of what was already in the library

Capture corrections and transferable insights. Leave durable facts to the research library and rules to the static context.

Modes

Mode	Behavior	Use for
ALWAYS	Start extraction from the current input on every run	Steady accumulation of observations
AGENTIC	The agent decides what is worth keeping	Research, where signal-to-noise matters
PROPOSE	The model proposes a learning and asks before saving; approval is prompt-based	Low-risk review where soft approval is sufficient

This is the same machine the data agent uses to self-correct, pointed at a shared store instead of a per-warehouse one.

Next steps

Task	Guide
Feed it grounded context too	Grounding research
Audit what changed the team’s mind	Structured deliverable

Developer Resources

• Learning Machines
• Learning modes
• Learned knowledge cookbook