Handoff input filter

Skill kjuhwa/skills-hub/skills/llm-agents/handoff-input-filter

Self-correcting knowledge corpus for Claude Code — 9 stable shape clusters, bias-correction pipeline baked into contribution flow. 47 papers, 45 techniques, 1.1k skills.

Install

npx -y skills add kjuhwa/skills-hub --skill handoff-input-filter

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

Filter or transform the conversation history passed to a specialist agent on handoff.

SKILL.md

2.5 KB, 439 tokens by cl100k_base, as published. Nobody here has run it

handoff-input-filter

Pass input_filter to handoff() to control what conversation history the specialist agent receives. Use built-in filters from agents.extensions.handoff_filters or write custom ones.

When to apply

When specialist agents need a clean context (e.g., remove tool calls, trim early history, remove PII) rather than the full accumulated conversation from the triage agent.

Core snippet

from agents import Agent, HandoffInputData, Runner, handoff, trace
from agents.extensions import handoff_filters

def my_handoff_filter(data: HandoffInputData) -> HandoffInputData:
    # Remove all tool-related messages from history
    data = handoff_filters.remove_all_tools(data)
    
    # Optionally trim early history
    history = (
        tuple(data.input_history[2:])
        if isinstance(data.input_history, tuple)
        else data.input_history
    )
    return HandoffInputData(
        input_history=history,
        pre_handoff_items=tuple(data.pre_handoff_items),
        new_items=tuple(data.new_items),
    )

specialist = Agent(
    name="Spanish Assistant",
    instructions="You only speak Spanish and are extremely concise.",
    handoff_description="A Spanish-speaking assistant.",
)

triage = Agent(
    name="Assistant",
    instructions="If the user speaks Spanish, handoff to the Spanish assistant.",
    handoffs=[handoff(specialist, input_filter=my_handoff_filter)],
)

Built-in filters

from agents.extensions import handoff_filters
handoff_filters.remove_all_tools(data)       # Strip tool calls and outputs
handoff_filters.keep_last_n_items(data, 5)   # Keep only recent items

Key notes

HandoffInputData has .input_history, .pre_handoff_items, .new_items, .run_context
Use .clone(**kwargs) to create a modified copy
Filters run synchronously; keep them fast
nest_handoff_history on handoff() controls whether history is nested under the new agent's turn

What ships with it

Read from the repository

Just SKILL.md. No reference files, no scripts.