skills.rest
BerriAI
BerriAIOFFICIAL
Explore all skills by this author →
WebsiteGitHub

update-agent

Update AI agent configurations on a LiteLLM proxy via PATCH requests.

77|12|Updated Mar 14, 2026
Search Tags:#api#agent#mcp#update#patch#proxy#liteLLM
One-click install
npx skills add https://github.com/BerriAI/litellm-skills --skill update-agent-berriai
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: update-agent
Source: https://github.com/BerriAI/litellm-skills/tree/main/update-agent
Command: npx skills add https://github.com/BerriAI/litellm-skills --skill update-agent-berriai

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Update an existing AI agent on a live LiteLLM proxy to modify its model, description, or MCP server access without re-creating the agent. This enables operational changes to agent capabilities and governance while preserving the agent_id.

Core Features & Use Cases

  • Update the underlying model assigned to an agent.
  • Change the agent description for clearer governance and documentation.
  • Modify MCP server access, TPM/RPM limits, or related agent policies.
  • Validate agent_id and confirm changes via the API response.

Quick Start

Identify the agent_id, specify the fields to update (model, description, tpm_limit, rpm_limit, or MCP servers), then send a PATCH request to /v1/agents/{agent_id}.

Frequently Asked Questions about update-agent

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I update an AI agent's configuration on a live LiteLLM proxy?▼

To update an AI agent on a LiteLLM proxy, send a PATCH request to the /v1/agents/{agent_id} endpoint with a JSON payload specifying new model, description, or rate limit fields. This requires a valid proxy admin key and curl.

Can I change MCP server access for an existing agent without re-creating it?▼

Yes, you can modify MCP server access for an existing agent by sending a PATCH request to the LiteLLM proxy API. This allows you to update agent capabilities and governance policies while preserving the original agent_id.

What fields can I modify when patching an agent on a LiteLLM proxy?▼

When patching an agent on a LiteLLM proxy, you can modify the assigned model, agent description, TPM limits, RPM limits, and MCP server access. Specify these fields in your JSON payload to apply the operational changes.

Do I need a proxy admin key to send PATCH requests to the LiteLLM agents endpoint?▼

Yes, a valid proxy admin key is required to authenticate PATCH requests sent to the /v1/agents/{agent_id} endpoint on a LiteLLM proxy. You also need curl installed to execute the request and validate the API response.

Why update an agent's model instead of deleting and re-creating the agent?▼

Updating an agent's model via PATCH preserves the existing agent_id, ensuring continuity for dependent systems. This method allows operational changes to agent capabilities, rate limits, and MCP server access without the disruption of recreating the agent.

Related Skills

Explore More Software Engineering Skills→
BerriAIBerriAI
77

delete-agent

Delete an AI agent from a live LiteLLM proxy via the admin API.

Official
Intermediate
BerriAIBerriAI
77

add-team

Create a new team on a LiteLLM proxy via POST /team/new.

Official
Intermediate
BerriAIBerriAI
77

add-mcp

Register MCP servers on a live LiteLLM proxy via POST /v1/mcp/server.

Official
Intermediate
BerriAIBerriAI
77

add-agent

Create an AI agent on a LiteLLM proxy via POST /v1/agents.

Official
Intermediate