Catalog
BerriAI/update-model

BerriAI

update-model

Update an existing model on a live LiteLLM proxy. Ask for the model_id and what to change (API key, base URL, etc.), then call POST /model/update.

v1.0LATEST
NewUpdated Oct 12, 2026

Update Model

Update an existing model's configuration on a live LiteLLM proxy.

Setup

LITELLM_BASE_URL  — e.g. https://my-proxy.example.com
LITELLM_API_KEY   — proxy admin key

Ask the user

  1. model_id (required) — if they don't have it, list models first:
    curl -s "$BASE/model/info" -H "Authorization: Bearer $KEY" | python3 -c "
    import sys,json
    for m in json.load(sys.stdin).get('data',[]):
      print(m['model_info']['id'], m['model_name'])
    "
    
  2. What to change — any combination of:
    • api_key (rotate the credential)
    • api_base (change endpoint)
    • api_version (Azure)
    • model (underlying model string, e.g. azure/new-deployment)

Run

curl -s -X POST "$BASE/model/update" \
  -H "Authorization: Bearer $KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model_info": {"id": "<model_id>"},
    "litellm_params": {
      "api_key": "<new_key>",
      "api_base": "<new_base>"
    }
  }'

Only include litellm_params fields being changed.

Output

Confirm the model was updated. Offer to run a test call to verify it still routes correctly.

Files1
1 files · 1.0 KB

Select a file to preview

Overall Score

72/100

Grade

B

Good

Grades are signals, not a certification. Always review a skill yourself before use.

Safety

70

Quality

68

Clarity

80

Completeness

65

Summary

This skill enables an AI agent to update model configurations on a live LiteLLM proxy server via REST API. The agent collects a model ID and desired changes (API key, base URL, etc.) from the user, then submits a POST request to the /model/update endpoint with the new configuration in JSON format.

Detected Capabilities

curl HTTP requestsJSON parsing (python3)environment variable reads (LITELLM_BASE_URL, LITELLM_API_KEY)POST API callscredential handling

Trigger Keywords

Phrases that agents use to match this skill to user intent.

update litellm modelrotate proxy keychange api endpointlitellm configurationmanage proxy models

Risk Signals

WARNING

Environment variable LITELLM_API_KEY contains proxy admin key

Setup section
WARNING

Admin credentials passed as Bearer token in Authorization header

curl examples
INFO

api_key field in litellm_params accepts arbitrary new credential

Run section — litellm_params.api_key
WARNING

No validation shown for model_id or api_key inputs before submission

Run section
INFO

Proxy endpoint domain (my-proxy.example.com) is trusted but unauthenticated HTTPS not verified in curl examples

curl invocations

Referenced Domains

External domains referenced in skill content, detected by static analysis.

my-proxy.example.com

Use Cases

  • Rotate API keys for proxied models without redeploying
  • Switch model endpoints or change Azure API versions for active deployments
  • Update underlying model strings (e.g. switch Azure deployments) on live proxies
  • List available models before selecting one to update

Quality Notes

  • Clear scope: only updates existing models on a specific proxy, not provisioning or deletion
  • Well-structured: Setup → Ask → Run → Output flow is logical and easy to follow
  • Helpful context: includes command to list models before update, reducing user friction
  • Good restriction: only operates on explicitly provided litellm_params fields (explicit over implicit)
  • Documentation mentions test verification but does not provide a concrete test call example
  • No error handling guidance: does not document how to handle 401 (bad key), 404 (model not found), or JSON malformation responses
  • Missing validation: no guidance on sanitizing model_id or validating new api_key format before submission
  • Endpoint assumption: assumes /model/info and /model/update exist on the proxy without version or discovery guidance
Model: claude-haiku-4-5-20251001Analyzed: Oct 12, 2026

Reviews

Add this skill to your library to leave a review.

No reviews yet

Be the first to share your experience.

Use BerriAI/update-model in your dev environment

Command Palette

Search for a command to run...