Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →yogsoth-ai/adversarial-escalation
Strategy: Progressive pressure escalation — starts with surface-level challenges and escalates to fundamental assumption attacks based on defender confidence decay.
Passed 19 of 21 test cases with the skill, and 6 of 21 without it.
and 19 more in the full report
15 fixed5 passed either way1 still failing1 broke
Tokens, turns and time describe resource use, not a dollar price.
One of 14 test cases the skill fixed. Same model, same prompt, with and without the skill.
What specific subagent is responsible for measuring defender resilience between rounds and acting as the escalation trigger in our automated policy debate pipeline?
| Checked | Without skill | With skill |
|---|---|---|
| The response identifies the confidence calibration module as evaluating defense resilience to trigger escalation or termination. | Missed | Passed |
The instructions your model reads when the skill loads.
Progressive pressure: escalate attack sophistication based on defender performance.
| Parameter | S | M | L |
|---|---|---|---|
| Debate rounds | 4 | 8 | 12 |
| Participating agents | 3 | 5 | 8 |
| Coverage dimensions | 3 | 5 | 7 |
| External evidence searches | 2 | 5 | 10 |
debate-architect → [design escalation ladder]
→ [for each level]:
debate-critic (level-appropriate attack)
→ debate-defender → debate-judge
→ confidence-calibration
→ (escalate if survived, terminate if collapsed)
→ debate-transcript-analysis → verdict-synthesis
test cases passed, vs 6 / 21 without
Published skills whose descriptions closely match this one, with their headline benchmark lift.
More skills from the same source and nearby in the registry, with their headline benchmark lift.