Skip to main content
Tool Config controls how tools run — their timeout, retry behavior, output limits, and whether large results get stored as artifacts. 
from praisonaiagents import Agent
from praisonaiagents.config.feature_configs import ToolConfig

agent = Agent(
    name="DataAgent",
    instructions="You analyze data using tools.",
    tools=[my_data_tool],
    tool_config=ToolConfig(
        timeout=60,
        parallel=True,
        enable_artifacts=True,
    )
)

result = agent.start("Analyze the sales data and generate a report")
print(result)
The user asks the agent to use tools; ToolConfig controls timeouts, retries, parallel execution, and artifact storage for those calls.

​
Quick Start

1

Simple Usage

Use ToolConfig() defaults to get retry protection and safe output limits:
from praisonaiagents import Agent
from praisonaiagents.config.feature_configs import ToolConfig

agent = Agent(
    name="Assistant",
    instructions="You are a helpful assistant with tools.",
    tools=[search_tool],
    tool_config=ToolConfig()
)

agent.start("Search for recent AI news")
2

With Custom Settings

Configure timeout, retries, and parallel execution:
from praisonaiagents import Agent
from praisonaiagents.config.feature_configs import ToolConfig

agent = Agent(
    instructions="You are a multi-source research agent.",
    tool_config=ToolConfig(
        parallel=True,
        timeout=60,
        output_limit=32000,
    ),
)
agent.start("Get weather, news, and stock data for San Francisco.")
3

With Artifact Storage

from praisonaiagents import Agent
from praisonaiagents.config.feature_configs import ToolConfig

agent = Agent(
    instructions="You are a data analysis agent.",
    tool_config=ToolConfig(
        enable_artifacts=True,
        artifact_retention_days=14,
        output_limit=32000,
    ),
)
agent.start("Analyze this large CSV file and summarize the findings.")

​
How It Works

PhaseWhat happens
1. ExecuteTool runs within the configured timeout
2. RetryFailed executions retry with exponential backoff
3. OutputLarge outputs spill to artifact storage
4. RespondAgent uses tool output to form the final answer

​
Configuration Options

ToolConfig SDK Reference

Full parameter reference for ToolConfig
Precedence ladder: 
# Level 1: Bool (enable with defaults)
agent = Agent(tool_config=True)

# Level 2: ToolConfig (full control)
agent = Agent(tool_config=ToolConfig(
    timeout=60,
    parallel=True,
    enable_artifacts=True,
))
OptionTypeDefaultDescription
timeoutint | NoneNonePer-tool timeout in seconds
retry_policyRetryPolicy | NoneNoneRetry with exponential backoff
parallelboolFalseRun batched LLM tool calls concurrently
output_limitint16000Max bytes before spilling to artifact store
output_max_linesint | NoneNoneMax lines before spilling
output_directionstr"both"Truncation direction: "head", "tail", or "both"
enable_artifactsboolFalseEnable artifact storage for large outputs
artifact_retention_daysint7Days to keep artifacts before cleanup
artifact_storeAny | NoneNoneCustom artifact store instance
redact_secretsboolTrueRedact secrets from artifacts
Legacy keyword arguments for tools now raise TypeError. Always use ToolConfig for tool configuration. See the migration guide if upgrading from an older version.

​
Common Patterns

​
Pattern 1 — Timeout for slow external APIs

from praisonaiagents import Agent
from praisonaiagents.config.feature_configs import ToolConfig

agent = Agent(
    name="DataAnalyzer",
    instructions="Analyze large datasets and return summaries.",
    tools=[run_sql_query],
    tool_config=ToolConfig(
        output_limit=32000,          # Larger inline limit
        enable_artifacts=True,       # Store huge outputs as files
        artifact_retention_days=14,  # Keep for 2 weeks
    )
)

agent.start("Run the monthly sales report query")
Parallel tool execution: 
from praisonaiagents import Agent
from praisonaiagents.config.feature_configs import ToolConfig

agent = Agent(
    name="ResearchAgent",
    instructions="Search multiple sources simultaneously.",
    tools=[search_news, search_arxiv, search_wikipedia],
    tool_config=ToolConfig(
        parallel=True,   # Run all three searches at once
        timeout=15,      # Fail fast if any tool is slow
    )
)

agent.start("Research the current state of quantum computing")
Tail-only output for log tools: 
from praisonaiagents import Agent
from praisonaiagents.config.feature_configs import ToolConfig

agent = Agent(
    name="LogAnalyzer",
    instructions="Check system logs for errors.",
    tools=[get_system_logs],
    tool_config=ToolConfig(
        output_direction="tail",  # Keep the most recent log lines
        output_max_lines=500,
    )
)

agent.start("Check recent server error logs for anomalies.")

​
Best Practices

Any tool that calls an external API or runs a subprocess should have a timeout. Without one, a hung tool call blocks your agent indefinitely. Start with 30–60 seconds and adjust based on your tool’s expected latency.
Use parallel=True when your agent commonly calls multiple tools at once and those tools don’t depend on each other’s outputs. This can cut wall-clock time significantly.
If your tools return large datasets (SQL queries, file reads, API responses), set enable_artifacts=True. This prevents large outputs from filling the LLM context window, which wastes tokens and can cause errors.
Leave redact_secrets=True (the default) to prevent API keys, passwords, and tokens from being stored in artifact files or shown in tool outputs.

Artifact Storage

How artifacts are stored and retrieved

Async Tool Safety

Safe concurrent tool execution

Toolsets

Group and manage tools as sets

Tool Retry Policy

Configure retry behavior for failing tools