🪢 Open source agent evals & observability: Trace, evaluate, and improve LLM applications with one open platform.
TypeScript analyticsautogenevaluationlangchainlarge-language-modelsllama-indexllmllm-evaluationllm-observabilityllmopsmonitoringobservabilityopen-sourceopenaiplaygroundprompt-engineeringprompt-managementself-hostedycombinator
/**
* MCP Tool: updatePromptLabels
*
* Updates labels for a specific prompt version.
* Write operation with destructive hint.
* This is the ONLY way to modify existing prompts (labels only).
*/
import { z } from "zod";
import { defineTool } from "../../../core/define-tool";
import { ParamPromptName, ParamNewLabels } from "../validation";
import { updatePromptLabelsForApi } from "@/src/features/prompts/server/prompt-api-service";
import { buildPromptUrl } from "@langfuse/shared/src/server";
import { runMcpTool } from "../../../core/run-mcp-tool";
import { PROMPT_NAME_MAX_LENGTH } from "@langfuse/shared";
/**
* Base schema for JSON Schema generation (MCP client display)
* Uses simple types without refinements
*/
const UpdatePromptLabelsBaseSchema = z.object({
name: z
.string()
.min(1)
.max(PROMPT_NAME_MAX_LENGTH)
.describe("The name of the prompt"),
version: z.coerce
.number()
.int()
.positive()
.describe("The version number to update (required)"),
newLabels: z
.array(z.string())
.describe(
"Array of new labels to assign to the prompt version (can be empty to remove all labels). The 'latest' label is auto-managed and cannot be supplied.",
),
});
/**
* Input schema for runtime validation
* Uses full validation schemas with refinements
*/
const UpdatePromptLabelsInputSchema = z.object({
name: ParamPromptName,
version: z.coerce
.number()
.int()
.positive()
.describe("The version number to update (required)"),
newLabels: ParamNewLabels,
});
/**
* updatePromptLabels tool definition and handler
*/
export const [updatePromptLabelsTool, handleUpdatePromptLabels] = defineTool({
name: "updatePromptLabels",
description: [
"Update labels for a specific prompt version.",
"",
"Important:",
"- ONLY way to modify existing prompts (labels only)",
"- Specified labels are added to the version (preserving others not mentioned)",
"- Labels are unique across versions - setting a label on one version automatically removes it from others",
"- 'latest' label is auto-managed and cannot be set manually",
"- Only add the 'production' label when the user explicitly asks to promote this prompt version to production",
"- Cannot modify prompt content, type, or tags - use createTextPrompt or createChatPrompt for new versions",
"",
"Accepts: name, version (required), newLabels (array, can be empty to remove all labels)",
].join("\n"),
action: "prompts:CUD",
baseSchema: UpdatePromptLabelsBaseSchema,
inputSchema: UpdatePromptLabelsInputSchema,
handler: async (input, context) => {
return await runMcpTool({
spanName: "mcp.prompts.update_labels",
context,
attributes: {
"mcp.prompt_name": input.name,
"mcp.prompt_version": input.version,
"mcp.new_labels_count": input.newLabels.length,
},
fn: async () => {
const { name, version, newLabels } = input;
const { updatedPrompt } = await updatePromptLabelsForApi({
context,
ctx: context.auth,
promptName: name,
promptVersion: version,
newLabels,
});
// Return formatted response
return {
id: updatedPrompt.id,
name: updatedPrompt.name,
version: updatedPrompt.version,
labels: updatedPrompt.labels,
url: buildPromptUrl({
projectId: context.projectId,
name: updatedPrompt.name,
version: updatedPrompt.version,
}),
message: `Successfully updated labels for '${updatedPrompt.name}' version ${updatedPrompt.version}. Labels are now: ${updatedPrompt.labels.length > 0 ? updatedPrompt.labels.join(", ") : "(none)"}`,
};
},
});
},
destructiveHint: true,
});