Confirm a harmless shell request is still blocked during a controlled Dry Run, then render a non-persistent Markdown summary.
Loading...
You will perform a controlled Dry Run, confirm that a harmless run_shell request remains blocked by policy, and generate a Markdown result that is not persisted.
This guide requires Pro. If Mqttable returns locked, stop and review Pricing. Stop if it is unavailable. Dry Run must not execute any shell command, and the Markdown output must remain non-persistent.
Send this single paragraph to the Agent connected to Mqttable:
Main prompt
Use the connected Mqttable MCP tools, when Pro is available, to perform a controlled Dry Run proving a harmless run_shell request remains policy-blocked, then generate a non-persistent Markdown summary. Do not execute any shell command.
The Agent chooses temporary names, creates only isolated test objects, leaves existing configuration unchanged, and returns the exact object list in its completed report for cleanup.
After a successful run, confirm these signals in Mqttable:
The MCP result dashboard records the controlled request without applying a change or retaining a plan; the Markdown report remains non-persistent.
If any success signal is missing, stop without expanding the scope and use Troubleshooting.
In the same Agent conversation, send:
Cleanup prompt
Discard only the temporary Dry Run result listed in your completed report. Do not persist the Markdown summary or change anything else; confirm no temporary object remains.
Continue with Troubleshooting when an Agent task stops at a safety boundary.