AI agents call get_work_item to retrieve information from Ado without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
The tool name 'get_work_item' indicates a retrieval operation that queries work item data without modification or side effects. Despite the empty description lowering confidence slightly, the consistent naming pattern across the ado-mcp server (where sibling tools include create_work_item, delete_work_item for Write/Destructive, and extract_pipeline_run_data for Read) strongly suggests this is a read-only data…
From the tool's definition Tool name is 'get_work_item' which follows Read operation naming conventions (get/fetch/retrieve). No description provided, but the pattern matches other read operations in the catalog like 'list projects' and 'view logs' mentioned in the server description.
Attacks that exploit this kind of access
get_work_item. It is categorised as a Read tool in the Ado MCP Server, which means it retrieves data without modifying state.
Register the Ado MCP server in PolicyLayer and add a rule for get_work_item: 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 Ado. Nothing to install.
get_work_item is a Read tool with low risk. Read-only tools are generally safe to allow by default.
Yes. Add a rate_limit block to the get_work_item 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 get_work_item. 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.
get_work_item is provided by the Ado MCP server (raboley/ado-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Every MCP server has a record like this.
Type a name, get the same breakdown: verified identity, auth posture, risk grade, capabilities, recommended policy.
Teams ship this data inside their own products. See what a licence covers →