Installs into .claude/skills of the current project.
Are you the author of Perspective Critic?
Add the live security badge to your README. It updates with every re-scan.
[](https://www.skillsdirectory.com/skills/yogsoth-ai-perspective-critic)
---
name: perspective-critic
description: Evaluates artifact from a specific assigned perspective. Produces assessment grounded in that viewpoint's values, priorities, and expertise.
execution: subagent
prompt: ./prompt.md
input: artifact (string), perspective (string), other_perspectives_output (string)
used-by: [multiagent-debate]
---
# Perspective Critic
Evaluates artifact from a specific assigned perspective.
## Execution
Subagent — spawned via subagent-spawning/spawn-agent.
## Why Subagent
Each perspective requires dedicated context to maintain viewpoint consistency. Mixing perspectives in one context leads to averaging rather than genuine diversity.
## Input
- **artifact**: The artifact being evaluated
- **perspective**: Assigned viewpoint (e.g., "methodological skeptic", "domain expert", "practitioner")
- **other_perspectives_output**: Previous outputs from other perspectives (empty in round 1)
## Output
- **assessment**: Structured evaluation from this perspective
- **concerns**: Issues visible from this viewpoint
- **strengths**: Positives visible from this viewpoint
- **confidence**: How viable the artifact appears from this perspective (0.0–1.0)
## Budget
One unit = one perspective evaluation per round.
