Installs into .claude/skills of the current project.
Are you the author of Judge Verdict?
Add the live security badge to your README. It updates with every re-scan.
[](https://www.skillsdirectory.com/skills/yogsoth-ai-judge-verdict)
---
name: judge-verdict
description: Render an impartial verdict on advocate case vs critic attacks with explicit reasoning.
execution: subagent
prompt: ./prompt.md
input: advocate_case, critic_attacks
used-by: [steel-manning, validation]
---
# Judge Verdict
Renders an impartial verdict after weighing the advocate's case against the critic's attacks. The judge must address every argument explicitly and produce a clear ACCEPT/REJECT/REVISE decision with conditions.
## Execution
Spawns a subagent with the judge role. The subagent receives both the advocate case and critic attacks, then renders a verdict with full reasoning.
## Why Subagent
- Impartiality requires isolation from both advocate and critic roles
- Judge must weigh evidence without prior commitment to either side
- Separation ensures the verdict is based on argument quality, not role loyalty
## HARD-GATE
Output must include:
- Explicit verdict: ACCEPT, REJECT, or REVISE
- Point-by-point response to every critic attack
- Assessment of advocate case strength post-attack
- Conditions for acceptance (if ACCEPT or REVISE)
- Clear reasoning chain from evidence to verdict
