create_action
Create a new server-side Action on a published app. Actions let browser code call external APIs without exposing keys — secrets live server-side and are templated in at invoke time. Template syntax (IMPORTANT, do not mix up): {name} for runtime variables from the body (lowercase name, no dollar s...
This record as markdown: /tools/deloc-mcp/create-action.md
What create_action does on Deloc
AI agents use create_action to create or update resources in Deloc, usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Deloc environment.
Why create_action is rated Medium
This tool creates new data structures (server-side Actions) on a published app that modify the app's behavior and enable external API calls with credentials. This is a Write operation because it creates/adds new reversible configurations. Severity is high because misconfigured Actions could expose API credentials, call unintended endpoints, or enable unauthorized external operations.
From the tool's definition Tool description states 'Create a new server-side Action on a published app' and handles secrets/API credentials server-side with templating. The tool creates new persistent server-side configurations that can call external APIs.
Attacks that exploit this kind of access
The rule that runs create_action safely
PolicyLayer is an MCP gateway: it sits between your AI agents and Deloc, and checks every tool call against a rule you set before the call runs. Nothing changes on the server itself. For create_action, this is the rule to start with:
create_action stays usable, but capped: an agent stuck in a loop can't make hundreds of changes a minute. Everything else on the server is denied unless you say otherwise.
The button opens the PolicyLayer dashboard: create your workspace, connect Deloc, apply this rule, and every create_action call is checked against it from then on.
Questions about create_action
Create a new server-side Action on a published app. Actions let browser code call external APIs without exposing keys — secrets live server-side and are templated in at invoke time. Template syntax (IMPORTANT, do not mix up): {name} for runtime variables from the body (lowercase name, no dollar sign — e.g. {placement_id}), ${NAME} for stored secrets (uppercase name only — e.g. ${API_KEY}), {{viewer.email}} for trusted server-injected context. Walk the user through: (1) target URL + method, (2) any secrets they need (uppercase ${NAME}), (3) variables the browser will pass (lowercase {name}), (4) which roles can invoke. Then prompt them to set_action_secret for every ${SECRET} used, then test_action. It is categorised as a Write tool in the Deloc MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Deloc MCP server in PolicyLayer and add a rule for create_action: allow, deny, rate-limit, or require approval. Point your MCP client at the PolicyLayer proxy URL and the rule is enforced on every call, before it reaches Deloc. Nothing to install.
create_action is a Write tool with medium risk. Write tools should be rate-limited to prevent accidental bulk modifications.
Yes. Add a rate_limit block to the create_action rule in your PolicyLayer policy. For example, setting max: 10 and window: 60 limits the tool to 10 calls per minute. Rate limits are tracked per agent session and reset automatically.
Set action: deny in the PolicyLayer policy for create_action. The AI agent will receive a policy violation error and cannot call the tool. You can also include a reason field to explain why the tool is blocked.
create_action is provided by the Deloc MCP server (delocdev/deloc-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
More on Deloc, and thousands of servers like it.
This server
Across the catalogue