Evaluate with evidence

Track for prompt injection: red-team pass rate, false positive rate on benign, privileged action rate, time-to-contain. Report fractions like 36/40, not vague quality adjectives. Segment by language, plan tier, document length, or other slices that matter for the tool-using support agent.

Freeze inputs and neighboring versions while evaluating prompt injection. Change one control. Pair results case by case on the tool-using support agent. Define hard gates (severe errors, privacy, latency) before the bake-off. Use deterministic checks where possible; humans for nuance; model judges only with calibration against gold.

Numeric reminder for prompt injection: risk ≈ probability of successful injection × impact of available capability; reducing tool privilege cuts impact even when detection is imperfect

Resist adding a twelfth metric before the first three for prompt injection on the tool-using support agent have owners. This workload improves faster when a small scorecard is trusted than when a warehouse of unused plots exists.

Also pin one numeric memory from this prompt injection chapter: risk ≈ probability of successful injection × impact of available capability; reducing tool privilege cuts impact even when detection is imperfect That number is not decoration; it is a template for how claims about prompt injection on the tool-using support agent should look in design docs. Scoped specifically to prompt injection / tool-using support agent / evaluation.

People confuse prompt injection with neighboring buzzwords when debugging the tool-using support agent. Before changing prompts, ask whether the broken stage was evidence gathering, the prompt injection judgment itself, validation, or the product action. Fixing the wrong stage creates folklore (“we tried prompt injection and it failed”) that blocks the next team on the tool-using support agent. Scoped specifically to prompt injection / tool-using support agent / evaluation.

Write a five-line artifact for this page: goal, inputs, check, owner, stop rule. Invent one fluent failure that the check would catch. Keep details specific to prompt injection rather than generic AI advice.

Write a five-line artifact for this page: goal, inputs, check, owner, stop rule. Invent one fluent failure that the check would catch. Keep details specific to prompt injection rather than generic AI advice.

Write a five-line artifact for this page: goal, inputs, check, owner, stop rule. Invent one fluent failure that the check would catch. Keep details specific to prompt injection rather than generic AI advice.

Write a five-line artifact for this page: goal, inputs, check, owner, stop rule. Invent one fluent failure that the check would catch. Keep details specific to prompt injection rather than generic AI advice.