[Forces] Changes in execution
Problem to solve
When running our agent, validation is performed only on the selected root when either the Static or Dynamic arguments are passed. However, when using the All argument, the targeted root is not validated. Instead, we check vulnerabilities that are not associated with any root. This could lead to false secure executions if users run the agent with the wrong root nickname.
Intended users
Agent users.
Permissions and Security
N/A.
Proposal
- If users don't define a root, the execution should be against all active roots, including validation over vulnerabilities that are not assigned to any root (Dynamic and All arguments).
- If users define a root, before executing the agent, we should validate that the root is active. If it's active, run the agent against that root, including validation over vulnerabilities that are not assigned to any root (Dynamic and All arguments).
- If it's not active, the execution should fail, informing the user that the selected root is not present nor active in the scope.
Test plan
Test scenarios could be the ones mentioned above.
Steps
-
Make sure that the code contributions checklist has been followed.
What does success look like, and how can we measure that?
Zero falsely secure agent executions