Clone an agent
Creates a new agent that mirrors the source agent’s behaviour but is owned by the calling identity. Creation is atomic — the clone is either fully created or no state is persisted.
What is copied from the source:
descriptionruntime(model configuration)backend(backend type + sandbox configuration)system_prompt,tools,subagents,skillsuser_timezone— inherited from the source agent- any leftover files outside the typed fields above (e.g.
docs/README.mdattached via the escapefilesfield)
What the caller can override (sent in the request body):
name— defaults to"<source name> (Copy)"permissions— defaults to the source’sidentitywithvisibility=userandtenant_access_level=read. When supplied the object fully replaces the default; PATCH-style merging does not apply.
What is always reset for the clone (you cannot override these):
- agent ID, agent version, owner, timestamps
What is never copied (start fresh on the clone):
extras(caller-defined metadata bag, e.g. icon customisations, template IDs)- caller-defined
configurablebag entries - sharing state, tokens, triggers, and schedules
Cloning a built-in system template returns 422.
Authorizations
LangSmith API key (lsv2_...).
Path Parameters
Source agent ID
Query Parameters
Comma-separated relations to hydrate: owner (resolve owner_id into an owner object) and/or files (raw file map). The legacy include_files=true boolean is a deprecated alias for include=files.
owner, files Body
Optional overrides for the clone
Response
Created
The caller's access level on this agent.
READ, RUN, WRITE "WRITE"
"2026-01-15T09:30:00Z"
"Researches topics and drafts summaries."
Raw file map. Returned only when include_files=true.
"11111111-1111-1111-1111-111111111111"
Deprecated: use system_prompt. Echoed alongside system_prompt with
the same value for backwards compatibility.
True when the caller owns this agent.
true
"Research Assistant"
"7f8e9d0c-1b2a-3948-5766-8594a3b2c1d0"
Opaque identifier for the agent's current revision.
"rev_3f9a2b1c"
Subagents parsed from subagents/<name>/AGENTS.md + subagents/<name>/tools.json.
Agent system prompt parsed from AGENTS.md.
"You are a helpful research assistant."
Tool config parsed from tools.json.
"2026-01-15T09:30:00Z"

