10 min
Attach an MCP server
Point Helios at a remote MCP server, choose which of its tools workflows may call, and use Helios itself over MCP from other clients.
5,000 free credits to start.
- 01
Add the server URL
From Settings, then MCP Servers, add a remote server by its HTTP endpoint. Choose whether it belongs to the organization or to one project.
- 02
Choose how it authenticates
Servers connect with a bearer token or an OAuth flow. Tokens are stored encrypted and attached at call time, the same way integration credentials are. Set a timeout for slow servers.
- 03
Review the tools it exposes
Helios lists every tool the server advertises, with its description and input schema. Read the list before you attach it to anything. This is the surface you hand your workflows.
- 04
Allow only the tools you want
The tool allowlist narrows the server to a named subset. A server with forty tools can contribute the three that matter to a task.
- 05
Attach it to a workflow and run it
Enable the server on a workflow, name the tools in the prompt, and run it by hand. The transcript shows each MCP tool call next to the integration calls.
- 06
Point another agent at Helios
Helios is also an MCP server. A client such as Claude Code can call `find_relevant_apis` and `call_api` to reach any connected integration through Helios. The same roles apply from that side.
Keep going
Next guides
10 min
Run your first workflow
Describe a task in plain words, give the agent one tool, and read the run. Ten minutes from an empty workspace to a transcript you can read end to end.
8 min
Use secrets and variables
Keep the values a task needs out of the prompt. Variables hold configuration. Secrets hold credentials. The agent never sees a secret in plain text.
7 min
Pin repositories as context
Pin a GitHub repository to a workflow or a chat. The agent starts every run with it.
Templates
Put it to work
Templates that use what this guide covers.
Engineering
Pull request reviewer
Reviews every pull request against your conventions and leaves inline comments only where it found something concrete.
Engineering
Incident postmortem drafter
Rebuilds the incident timeline from Slack and GitHub, names the likely change, and drafts the write-up in Notion.
Operations
Nightly data quality checks
Runs your invariants against production every night and emails the on-call analyst which rows broke and since when.
Try it in your workspace.
Start free, connect a tool, and write the first prompt in plain words.