BoldCreate by Syncfusion

AI coding automation for scheduled and triggered work

Put your AI development team on a routine

Automations start Cloud Agent tasks for your engineering team: every weekday morning, when a pull request opens or a check fails, or when your monitoring tool or tracker calls a webhook. Each run works in its own isolated sandbox, and code changes arrive as pull requests your team reviews and merges.

Schedules, code events, and webhooks10 built-in recipesEvery change arrives as a pull request
Find critical bugsAutomation · Cloud AgentOn
1WHENEvery day at 9:00 a.m.
2WHEREacme/billing
3DOLook for high-severity bugs in the last day’s commits and fix them…
4WITHMemoryOpen pull requests
5THENEach run becomes a task you can open
TASKSWaiting for the next trigger
Built by Syncfusion, serving 47,000+ organizations.25+ years of developer tools, now reimagined for the agent-first era.
47,000+Syncfusion customers
1,600+Syncfusion UI components
25+Years of Syncfusion developer tools
Syncfusion component licensing is separate.
How it works

How an automation goes from trigger to review

An automation is a Cloud Agent task with a trigger. You decide when it starts, where it works, and what it may change. Your team still reviews and merges every pull request it opens.

ANY TRIGGER STARTS A RUNRUNS DON’T STOP TO ASK
Task in your sidebar
01 / 05 · WHENAdd one or more triggers

A schedule, a code event, or a webhook. Add more than one, and any of them starts a run.

automation · Find critical bugsrunning
schedule every day at 9:00 a.m.
runs at most once an hour
Triggers

When an automation runs

Pick the moment the work should happen. Combine triggers when one isn’t enough, for example a weekly run plus a run whenever a check fails.

  • Schedules: every hour, day, weekday, or week, or a custom cron expression in your time zone.
  • Code events on GitHub, GitLab, or Gitea: pull requests opened, updated, or merged, comments, reviews, and failed checks.
  • Webhooks: your CI pipeline, Sentry, Datadog, PagerDuty, Linear, or any system that can send a POST request.
On a schedule● run startedSunMonTueWedThuFriSatEvery weekday at 9:00 a.m. · 0 9 * * 1-5
Code eventslisteningGitHubGitLabGiteaPull request openedPull request updated (pushed or edited)Pull request mergedComment on a pull requestReview submittedCheck failed on a pull requestOnly run if: base branch is main
WebhooklisteningPOST /automations/…/runX-Webhook-Secret: ••••••••{ "alert": "5xx on /invoices", "service": "billing" }
Recipes

Start from 10 built-in recipes

Each recipe fills in the trigger, the instructions, and what the automation may do. Choose where it works, edit anything you like, and save. It stays off until you turn it on.

On a scheduleRECIPE 01 / 10

Find critical bugs

STARTSEvery day
DOESLooks for high-severity bugs in the last day’s commits and fixes each one with a test that fails before the fix.
RETURNSPull request
Edit anything before you saveOff until you turn it on
Guardrails

Automated runs get the same sandbox, spending cap, and audit log

No one approves a spec before an automated run, so the controls are set when you save the automation and enforced on every run.

  • Each run gets its own isolated sandbox, like any Cloud Agent task.
  • Runs count against your spending cap and are written to the audit log.
  • Runs work as you through your git connection, and review comments post as your organization’s bot.
  • Memory carries across runs, so a recipe can skip bugs, alerts, and tickets it already handled.
  • Changes arrive as pull requests, so your team still reviews and merges them.
Fix failing CI · run 14Automation · runs unattended
  1. Trigger
  2. Sandbox
  3. Instructions
  4. Check
  5. Budget
  6. Audit log
  7. Result
Illustrative example, not a quote
Set one up

Write one sentence and BoldCreate drafts the automation

You don’t need to write cron expressions by hand. Describe the routine, and BoldCreate fills in the trigger, the repository, and the instructions for you to review.

DESCRIBE AN AUTOMATIONDescribe an automation|Draft it
  1. 01Describe it, or pick a recipeWrite what you want in a sentence and BoldCreate drafts the automation for you to check before it’s saved. You can also start from a recipe, start blank, or import JSON.
  2. 02Pick the repositoriesPick the repositories or the environment. Code events listen to the repositories you choose here.
  3. 03Try a test runRun it once and read the task it produces before it runs on its own.
  4. 04Turn it onNew automations stay off until you switch them on. Switch one off at any time.
  5. 05Open any run to see what happenedEach run is a task in your sidebar, marked as the automation’s. Open any run to see what the agent did.
Safeguards

Limits and filters for each trigger type

Each trigger type has its own limits, so a noisy pipeline or a busy repository doesn’t start more runs than your team can review. Filter code events by base branch, head branch, author, title, labels, or draft status.

APPLIES TOSAFEGUARD
SchedulesRun at most once an hour, in the time zone you pick.
Code eventsIgnore the automation’s own comments and pull requests, and pull requests from forks. Add conditions on base branch, head branch, author, title, labels, or draft status.
WebhooksAccept a POST with the automation’s key (Authorization: Bearer or X-Webhook-Secret), up to 200 KiB. Up to 10 runs an hour, one at a time. The same body twice within 10 minutes runs once.
Webhook bodyReaches the agent as data, not as instructions. You can also require the body to match a pattern before a run starts.
New automationsStay off until you turn them on. Use Test run to try one first.

Frequently asked questions

What is AI coding automation in the Cloud Agent?
An automation is a Cloud Agent task that starts on its own. You set when it runs, where it works, and what it may do. Each run becomes a task in your sidebar with a pull request, a review comment, or its findings for your team.
What can start an automation?
A schedule (every hour, day, weekday, or week, or a cron expression), a code event on GitHub, GitLab, or Gitea, or a webhook from your CI, monitoring tool, or ticket tracker, such as Jira or Linear. Slack and Teams triggers are coming soon.
Can I create automations in the Desktop App or the CLI?
No. Automations are a Cloud Agent feature. You create and manage them in the Cloud Agent, and every run works in its own cloud sandbox.
Do automated runs have the same controls as other cloud tasks?
Yes. Each run gets its own isolated sandbox, counts against your spending cap, and is written to the audit log. Changes arrive as pull requests, so your team still reviews and merges them.
Are automations included in my plan?
Yes. Automations are included in the BoldCreate plan at no extra charge. Each run is billed for AI and sandbox usage at provider cost plus the 10% platform fee, like any other task.

Start with one automation in your repository

Pick a recipe, choose the repository it works in, and try a test run. Turn it on when the result looks right, and review each pull request it opens.

$199 a month for 20 seats · AI and sandboxes at provider cost + 10% platform fee