Knowledge Erasure

Erasure clears an agent’s knowledge base between sessions, with clear errors when an adapter can’t support an operation.

from praisonaiagents import Agent

agent = Agent(name="privacy-aware", instructions="Answer, then clear context between sessions.", knowledge=["policy.md"])
agent.start("Summarise the policy")
agent.knowledge.reset()   # clears the knowledge base between sessions

Quick Start

Reset the knowledge base

from praisonaiagents import Agent

agent = Agent(name="assistant", instructions="Answer from the docs", knowledge=["policy.md"])
agent.start("Summarise the policy")

# Clear the knowledge base between sessions
agent.knowledge.reset()

Delete all scoped memories

# Delete everything for a given scope
agent.knowledge.delete_all(user_id="user-123")

Inspect change history

# History is only available on Mem0-backed knowledge
changes = agent.knowledge.history(memory_id="mem-abc")

How It Works

Knowledge erasure calls route to the underlying memory adapter. Unsupported operations fail loudly instead of raising AttributeError.

Operation	Behaviour when unsupported
`reset()`	Logs a `WARNING`, no-ops, and returns `False` (returns `True` on a real reset)
`delete_all()`	Raises `NotImplementedError`
`history(memory_id)`	Raises `NotImplementedError`

Adapter Capabilities

Support differs per adapter — verify against praisonaiagents/knowledge/adapters/ before relying on an operation.

Operation	Chroma	SQLite	Mongo	Mem0
`reset()`	✅	✅	⚠️ warning + no-op	✅
`delete_all()`	✅	✅	❌ `NotImplementedError`	✅
`history(id)`	❌ `NotImplementedError`	❌ `NotImplementedError`	❌ `NotImplementedError`	✅

Since PraisonAI PR #4020, Knowledge.reset() warns instead of raising AttributeError, and delete_all() / history() raise NotImplementedError on unsupported adapters instead of AttributeError. Wrap history() and delete_all() in try/except when your code should work across adapters.

Since PraisonAI PR #4991, Knowledge.reset() returns bool — True when the backend actually reset, False when the adapter has no reset() and the call was a silent no-op. On destructive paths, check the return value before telling a user their data was erased: a False means nothing was cleared. Existing callers that ignore the return value are unaffected.

Common Patterns

Guard erasure calls so your code works across adapters:

from praisonaiagents import Agent

agent = Agent(name="assistant", instructions="Answer from the docs", knowledge=["policy.md"])

# reset() never raises — it warns and no-ops on adapters without native reset
agent.knowledge.reset()

# delete_all() / history() may not be supported
try:
    agent.knowledge.delete_all(user_id="user-123")
except NotImplementedError:
    # Fall back to reset() on adapters without scoped delete
    agent.knowledge.reset()

Report to users honestly on a destructive path — check the return value before claiming success:

from praisonaiagents import Agent

agent = Agent(name="assistant", knowledge=["policy.md"])

# Never claim success on a destructive path without checking
if agent.knowledge.reset():
    print("Knowledge cleared")
else:
    backend = type(agent.knowledge.memory).__name__
    print(f"Knowledge was NOT cleared — backend {backend} has no reset()")

Best Practices

Reset between sessions

Call agent.knowledge.reset() at the end of a session to clear context for privacy-sensitive workflows. reset() returns True when the backend actually reset, and False when the adapter had no reset — it warns and no-ops instead. On destructive paths check the return value before telling a user their data was erased; a False means nothing was cleared.

Wrap delete_all() and history()

Both raise NotImplementedError on adapters that don’t support them. Wrap them in try/except so multi-adapter code degrades gracefully.

Use Mem0 for history

Only Mem0-backed knowledge implements history(memory_id). Choose Mem0 when you need an audit trail of memory changes.

Check logs after upgrade

reset() returns False and logs a warning when the adapter can’t reset. Prefer the return value over log-scraping — it’s a reliable signal that the tier was actually cleared (True) or silently skipped (False).

Knowledge Overview

Load files and documents into an agent’s knowledge base.

Knowledge Storage

Configure the storage backend for knowledge.

Getting Started

Install

Core Concepts

Guides

Features

Models

Databases

Observability

Memory

Telemetry & Analytics

Knowledge

RAG

Persistence

Tools

Other Features

Developers

Configuration

Best Practices

Getting Started (No Code)

Quick Start

How It Works

Adapter Capabilities

Common Patterns

Best Practices

Knowledge Overview

Knowledge Storage

​Quick Start

​How It Works

​Adapter Capabilities

​Common Patterns

​Best Practices

​Related

Knowledge Overview

Knowledge Storage

Quick Start

How It Works

Adapter Capabilities

Common Patterns

Best Practices

Related