Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →majiayu000/atomic-agents-agent-configuration
This skill should be used when the user asks to "create an agent", "configure AtomicAgent", "set up agent", "agent configuration", "AgentConfig", "ChatHistory", or needs guidance on agent initialization, model selection, history management, and agent execution patterns for Atomic Agents applications.
Passed 22 of 22 test cases with the skill, and 13 of 22 without it.
and 8 more in the full report
9 fixed13 passed either way0 still failing0 broke
Tokens, turns and time describe resource use, not a dollar price.
One of 9 test cases the skill fixed. Same model, same prompt, with and without the skill.
When switching user conversations in an active Atomic Agent instance, I want to clear all accumulated conversation turns without destroying or re-instantiating the agent object itself. What method should be invoked on the agent?
| Checked | Without skill | With skill |
|---|---|---|
| Clears conversation history by calling agent.reset_history(). | Missed | Passed |
The instructions your model reads when the skill loads.
The AtomicAgent is the core class for LLM interactions in the Atomic Agents framework. It handles structured input/output, conversation history, and system prompt management.
import instructor
import openai
from atomic_agents.agents.base_agent import AtomicAgent, AgentConfig
from atomic_agents.lib.components.system_prompt_generator import SystemPromptGenerator
from atomic_agents.lib.components.chat_history import ChatHistory
# 1. Create instructor-wrapped client
client = instructor.from_openai(openai.OpenAI())
# 2. Configure the agent
config = AgentConfig(
client=client,
model="gpt-4o-mini",
history=ChatHistory(),
system_prompt_generator=SystemPromptGenerator(
background=["You are an expert assistant."],
steps=["1. Analyze the input.", "2. Generate response."],
output_instructions=["Be concise and helpful."],
),
)
# 3. Create the agent with type parameters
agent = AtomicAgent[InputSchema, OutputSchema](config=config)
| Parameter | Type | Required | Description |
|---|---|---|---|
client | Instructor client | Yes | Instructor-wrapped LLM client |
model | str | Yes | Model identifier (e.g., "gpt-4o-mini") |
history | ChatHistory | No | Conversation history manager |
system_prompt_generator | SystemPromptGenerator | No | System prompt configuration |
test cases passed, vs 13 / 22 without
More skills from the same source and nearby in the registry, with their headline benchmark lift.