MCP Tool Reference Medium Risk

upsert-prompt

Create or update a prompt template

Part of the Phoenix MCP server. Enforce policies on this tool with Intercept, the open-source MCP proxy.

Arize-ai/phoenix Write

WHEN AI AGENTS USE THIS TOOL

AI agents use upsert-prompt to create or modify resources in Phoenix. Write operations carry medium risk because an autonomous agent could trigger bulk unintended modifications. Rate limits prevent a single agent session from making hundreds of changes in rapid succession. Argument validation ensures the agent passes expected values.

WHY ENFORCE A POLICY ON UPSERT-PROMPT

Without a policy, an AI agent could call upsert-prompt repeatedly, creating or modifying resources faster than any human could review. Intercept's rate limiting ensures write operations happen at a controlled pace, and argument validation catches malformed or unexpected inputs before they reach Phoenix.

RECOMMENDED POLICY

Write tools can modify data. A rate limit prevents runaway bulk operations from AI agents.

phoenix.yaml 
tools:
  upsert-prompt:
    rules:
      - action: allow
        rate_limit:
          max: 30
          window: 60

See the full Phoenix policy for all 19 tools.

DETAILS

Tool Name

upsert-prompt

Category

Write

MCP Server

Phoenix MCP Server

Risk Level

Medium

MORE PHOENIX TOOLS

add-dataset-examples Write
Add new examples to an existing dataset
add-prompt-version-tag Write
Add a tag to a specific prompt version
list-projects Read
List all projects with optional pagination
list-datasets Read
List all datasets in the Phoenix instance
get-dataset-examples Read
Fetch examples from a specific dataset
get-dataset-experiments Read
List experiments run on a dataset
list-experiments-for-dataset Read
List all experiments for a dataset ID
get-experiment-by-id Read
Get full experiment metadata and results

View all 19 tools →

SIMILAR WRITE TOOLS ON OTHER SERVERS

create_table Airtable
Create a new table with fields
update_table Airtable
Update a table's name or description
create_field Airtable
Add a new field to a table
update_field Airtable
Modify an existing field
create_record Airtable
Add a new record
update_record Airtable
Modify an existing record

RELATED READING

FREQUENTLY ASKED QUESTIONS

What does the upsert-prompt tool do?

Create or update a prompt template. It is categorised as a Write tool in the Phoenix MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.

How do I enforce a policy on upsert-prompt?

Add a rule in your Intercept YAML policy under the tools section for upsert-prompt. You can allow, deny, rate-limit, or validate arguments. Then run Intercept as a proxy in front of the Phoenix MCP server.

What risk level is upsert-prompt?

upsert-prompt is a Write tool with medium risk. Write tools should be rate-limited to prevent accidental bulk modifications.

Can I rate-limit upsert-prompt?

Yes. Add a rate_limit block to the upsert-prompt rule in your Intercept policy. For example, setting max: 10 and window: 60 limits the tool to 10 calls per minute. Rate limits are tracked per agent session and reset automatically.

How do I block upsert-prompt completely?

Set action: deny in the Intercept policy for upsert-prompt. The AI agent will receive a policy violation error and cannot call the tool. You can also include a reason field to explain why the tool is blocked.

What MCP server provides upsert-prompt?

upsert-prompt is provided by the Phoenix MCP server (Arize-ai/phoenix). Intercept sits as a proxy in front of this server to enforce policies before tool calls reach the server.

ENFORCE POLICIES ON PHOENIX

Open source. One binary. Zero dependencies.

VIEW ON GITHUB Full Phoenix policy →