Skip to main content
Glama

relay_runs_list

List recent workflow runs for debugging and reference, enabling efficient AI workflow orchestration by chaining multi-step LLM operations.

Instructions

List recent workflow runs for debugging and reference.

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
limitNoNumber of runs to return (default: 10, max: 50)

Implementation Reference

  • The core handler function for the 'relay_runs_list' tool. Fetches recent runs from the run store, computes summaries with metrics like duration, tokens, cost, and trace URLs, and returns the formatted list.
    export async function relayRunsList(
      input: RelayRunsListInput
    ): Promise<RelayRunsListResponse> {
      const limit = input.limit || 10;
      const runs = getRecentRuns(limit);
      const config = getConfig();
    
      const summaries: RunSummary[] = runs.map((run: RunRecord) => ({
        runId: run.runId,
        type: run.type,
        name: run.type === 'workflow' ? run.workflowName : undefined,
        model: run.type === 'single' ? run.model : undefined,
        success: run.success,
        startTime: run.startTime.toISOString(),
        durationMs: run.durationMs,
        totalTokens: run.usage.totalTokens,
        estimatedCostUsd: run.usage.estimatedProviderCostUsd,
        traceUrl: `${config.traceUrlBase}/${run.runId}`,
        contextReduction: run.contextReduction,
      }));
    
      return {
        runs: summaries,
        total: summaries.length,
      };
    }
  • Zod schema defining the input parameters for the relay_runs_list tool, including an optional limit for the number of runs.
    export const relayRunsListSchema = z.object({
      limit: z
        .number()
        .min(1)
        .max(50)
        .optional()
        .describe('Number of runs to return (default: 10, max: 50)'),
    });
  • MCP tool definition object specifying the name, description, and input schema structure for registration.
    export const relayRunsListDefinition = {
      name: 'relay_runs_list',
      description: 'List recent workflow runs for debugging and reference.',
      inputSchema: {
        type: 'object' as const,
        properties: {
          limit: {
            type: 'number',
            description: 'Number of runs to return (default: 10, max: 50)',
          },
        },
      },
    };
  • src/server.ts:59-67 (registration)
    Array of all tool definitions, including relayRunsListDefinition, registered with the MCP server's listTools request handler.
    const TOOLS = [
      relayModelsListDefinition,
      relayRunDefinition,
      relayWorkflowRunDefinition,
      relayWorkflowValidateDefinition,
      relaySkillsListDefinition,
      relayRunsListDefinition,
      relayRunGetDefinition,
    ];
  • Dispatch logic in the MCP callTool request handler: parses arguments with the tool schema and invokes the relayRunsList handler function.
    case 'relay_runs_list': {
      const parsed = relayRunsListSchema.parse(args || {});
      result = await relayRunsList(parsed);
      break;

Latest Blog Posts

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/RelayPlane/mcp-server'

If you have feedback or need assistance with the MCP directory API, please join our Discord server