Open the local, read-only SAP control panel: an HTML page with buttons for the read-only tools (search, grep source, get source, read table, ATC, where-used, packages, transports, dumps, inactive objects). Returns the URL to open. The panel is served from inside THIS MCP process, so it works only...
AI agents call adt_open_panel to retrieve information from Claude For Abap without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool opens a local, read-only control panel serving only read operations (search, grep, get source, read table, etc.). It explicitly states no write tools are reachable and is bound to localhost with a security token. The only side effect is opening a browser tab locally, which is negligible. This is fundamentally a Read operation with low blast radius.
From the tool's definition 'read-only SAP control panel', 'read-only tools only — no write tool is reachable', 'Bound to 127.0.0.1, gated by a per-boot random token'
Attacks that exploit this kind of access
Open the local, read-only SAP control panel: an HTML page with buttons for the read-only tools (search, grep source, get source, read table, ATC, where-used, packages, transports, dumps, inactive objects). Returns the URL to open. The panel is served from inside THIS MCP process, so it works only while this session keeps the MCP connected and dies when the session ends. Bound to 127.0.0.1, gated by a per-boot random token, read-only tools only — no write tool is reachable. By default also opens the URL in the local default browser. It is categorised as a Read tool in the Claude For Abap MCP Server, which means it retrieves data without modifying state.
Register the Claude For Abap MCP server in PolicyLayer and add a rule for adt_open_panel: 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 Claude For Abap. Nothing to install.
adt_open_panel 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 adt_open_panel 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 adt_open_panel. 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.
adt_open_panel is provided by the Claude For Abap MCP server (yzonur/claude-for-abap). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
adt_open_panel is one line of Claude For Abap's registry record.
The record carries the whole server: verified identity, auth posture, risk grade, every tool classified, recommended policy — re-checked continuously.
Teams ship this data inside their own products. See what a licence covers →