SuperAPI AI Directory

iblai-vibe-agent-task

Agents & models

View repository

Updated 10-15 22:16

About this skill

/iblai-vibe-agent-task

Add the agent Tasks tab -- schedule automated tasks that run your agent on a cron cadence (one-off, daily, weekly, or monthly), then track each run. The tab bundles four surfaces into one component: a searchable, date-filterable task list with status badges, metric cards (total / completed / failed), a Schedule Task dialog, a per-task run-logs panel, and a log-details modal. This is one tab in the wider agent-settings family (access, api, datasets, disclaimers, embed, history, llm, memory, prompts, safety, settings, tasks, tools). Each tab is a separate skill. All tabs share the same AgentSettingsProvider wrapper -- set it up once and mount as many tabs as you need.

Task list with metric cards and run-logs panel Schedule Task dialog (calendar, name, prompt, time, repeat, email) Task list with run statuses and the logs panel Log Details modal (status, timing, output)

Common setup (brand, conventions, env files, verification): see docs/skill-setup.md.

Prerequisites

  • Auth must be set up first (/iblai-vibe-auth)
  • MCP server + skills configured (@iblai/mcp in .mcp.json)
  • AgentSettingsProvider must wrap the route (see /iblai-vibe-agent-setting Step 2 if not already set up)
  • Ask the user for a real mentorId (agent UUID). Do NOT invent one.

Step 1: Check Environment

Before proceeding, check for an iblai.env in the project root. Look for PLATFORM, DOMAIN, and TOKEN variables. If the file does not exist or is missing these variables, tell the user: "You need an iblai.env with your platform configuration. Download the template and fill in your values: curl -o iblai.env https://raw.githubusercontent.com/iblai/vibe/refs/heads/main/iblai.env"

Step 2: Mount AgentTasksTab

// app/(app)/agents/[mentorId]/tasks/page.tsx
"use client";

import { AgentTasksTab } from "@iblai/iblai-js/web-containers/next";

export default function AgentTasksPage() {
  return (
    <div className="flex h-full flex-col bg-white">
      <AgentTasksTab />
    </div>
  );
}

AgentTasksTab reads tenantKey, mentorId, and username from AgentSettingsProvider (via useAgentSettings()) and handles all of its own data fetching and mutations (list, create, delete, run logs). No props are required.

With custom pagination

Both the task list and the run-logs panel paginate (5 per page). Without a PaginationComponent the built-in web-containers IblPagination is used. Inject your own to match the host app's pagination UI:

<AgentTasksTab
  PaginationComponent={({ currentPage, totalPages, onPageChange, disabled }) => (
    <MyPagination
      page={currentPage}
      total={totalPages}
      onChange={onPageChange}
      disabled={disabled}
    />
  )}
/>;

Step 3: Customize Labels (Optional)

AgentTasksTab renders with the default agent-facing copy (AGENT_TASKS_TAB_LABELS). Override any string via the labels prop. Pass a full TasksTabLabels bundle (for a full agent re-skin) or a partial object (for one-off edits).

import {
  AgentTasksTab,
  type TasksTabLabels,
  AGENT_TASKS_TAB_LABELS,
} from "@iblai/iblai-js/web-containers/next";

<AgentTasksTab
  labels={{
    header: {
      title: "Scheduled tasks",
      description: "Schedule automated runs for your agent.",
    },
    toolbar: { scheduleTask: "New task" },
  }}
/>;

The label bundle is grouped by surface: header, toolbar, metrics, list, logs, states, scheduleDialog, deleteDialog, logDetails, and toasts. Override only the keys you need.

Step 4: Use MCP Tools for Customization

get_component_info("AgentTasksTab")
get_component_info("AgentSettingsProvider")

Tab surfaces

AgentTasksTab is a single mounted component, but it renders four distinct surfaces. The corresponding web-containers source (under web-containers/.../edit-mentor-modal/tabs/tasks-tab) is listed for reference -- these pieces are internal to AgentTasksTab and are not mounted separately.

Surface Screenshot Source component What it shows
Task list list tasks-tab-content / task-list / task-metrics-cards / tasks-tab-toolbar Search + date filter + Schedule Task button, metric cards (Total / Completed / Failed), and a list of tasks with time, repeat cadence, and a status badge (Scheduled / Running / Completed / Failed / Disabled)
Schedule Task dialog create schedule-task-dialog Calendar date picker plus Task Name, Task Prompt, Time, Repeat (Don't repeat / Daily / Weekly / Monthly), and a "Notify me by email" toggle
Run-logs panel logs task-logs-panel Per-task run logs; click a task on the left to load its logs, click a log row to open its details
Log Details modal log details task-log-details A single run's status, Created / Started / Ended timestamps, and full output

Repeat cadence maps to a cron schedule: "Don't repeat" creates a one_off task pinned to the chosen date; Daily / Weekly / Monthly create a recurring cron. The list status is derived from the latest run log combined with the task's enabled flag -- a one-off that ran successfully shows Completed, not Disabled.

<AgentTasksTab> Props

Import from @iblai/iblai-js/web-containers/next.

Prop Type Required Description
PaginationComponent ComponentType<{ currentPage, totalPages, onPageChange, disabled, disableNumberedButtons? }> No Custom pagination for the task list and logs panel. Defaults to the web-containers IblPagination
labels DeepPartial<TasksTabLabels> No Override user-visible strings

Related Exports

From @iblai/iblai-js/web-containers/next:

  • AGENT_TASKS_TAB_LABELS -- the default agent-facing label bundle.
  • TasksTabLabels -- type for the full label bundle.
  • TaskListItem -- a periodic agent normalized for the list (id, name, time, repeat, oneOff, enabled, status, rawData).
  • TaskLog -- a run log normalized for the logs panel (id, entry, timestamp, periodicAgentId, status, startTime, endTime).
  • TaskDisplayStatus -- the derived status union (Disabled | Running | Failed | Completed | Scheduled).

Step 5: Verify

Run /iblai-vibe-ops-test before telling the user the work is ready:

  1. pnpm build -- must pass with zero errors
  2. pnpm test -- vitest must pass
  3. Start dev server and touch test:
    pnpm dev &
    npx playwright screenshot http://localhost:3000/agents/<id>/tasks /tmp/agent-tasks.png
    

Important Notes

  • Redux store: Must include mentorReducer and mentorMiddleware -- the periodic-agent API slice (list / create / delete / logs) ships inside those bundles.
  • initializeDataLayer(): 5 args (v1.2+)
  • @reduxjs/toolkit: Deduplicated via webpack aliases in next.config.ts
  • Peer deps: sonner and @iblai/iblai-web-mentor must be installed (pnpm add sonner @iblai/iblai-web-mentor)
  • Shared provider: AgentSettingsProvider must wrap the route at a layout level. See /iblai-vibe-agent-setting Step 2 for the full snippet.
  • One-off vs recurring: "Don't repeat" flags the task one_off; the backend runs it once and the scheduler auto-disables it afterward. The list still shows the last outcome (e.g. Completed), not Disabled.
  • Start time: The dialog combines the picked date and time into the task's start_time; it must be in the future or the dialog blocks save.
  • Brand guidelines: BRAND.md