Passed 17 of 20 test cases with the skill, and 10 of 20 without it.
Without skill50%
With skill86%
Checks it passed that the model alone missed
Searches Panorama NAT policy rules
Searches Panorama security policy rules
Uses python3 $MCP_CALL as the execution script
and 8 more in the full report
8 fixed11 passed either way3 still failing0 broke
Pass-rate lift
+36 pp
Tokens
-24%
Agent turns
0%
Response time
-52%
Model gemini-3.6-flashMeasured 8/16/2026v119 of 22 cases passed with skill · 20 comparable
Tokens, turns and time describe resource use, not a dollar price.
See it in action
One of 8 test cases the skill fixed. Same model, same prompt, with and without the skill.
1 of 8
Prompt
We need to check address objects defined in Panorama. Write out the bash command line syntax required to invoke an MCP tool named list_address_objects passing parameter device_group set to Branch-East using standard environment helper variables.
What was checked, without and with the skill
Checked
Without skill
With skill
Uses python3 $MCP_CALL as the execution script
Missed
Passed
Passes $PANOS_MCP_CMD as the target command argument
Missed
Passed
case-04 · gemini-3.6-flash · 8/16/2026 · answers are capped and redacted
What's inside
The instructions your model reads when the skill loads.