Self Eval
Honestly evaluate AI work quality using a two-axis scoring system.
by AgentVoltv1.0.0Published 1y ago1 install
Free to sign up · every skill included with AgentVolt Pro
Before & after
Same prompt, same agent — the only difference is the skill.
Prompt“Rate the quality of the social media posts I just created”
Beforegeneric AI agent
I'd be happy to give you feedback on your social media posts. To provide meaningful evaluation, what specific criteria matter most to you - engagement potential, brand consistency, tone and voice, visual appeal, or something else? What's the platform and audience? Are these posts meant to drive awareness, engagement, conversions, or community building? Do you want me to focus on particular aspects like copy length, hashtag strategy, or visual composition? Once I understand what success looks like for these posts, I can give you my thoughts on how well they hit those goals.
Afterwith Self Eval
SELF EVAL - SOCIAL MEDIA POST QUALITY ASSESSMENT
EVALUATION FRAMEWORK: Two-axis scoring (1-10 each)
X-axis: BRAND CONSISTENCY (voice, visual, positioning)
Y-axis: ENGAGEMENT POTENTIAL (shareability, comments, saves)
POST 1: "Quick productivity hack" (Instagram Reel script)
Brand consistency: 8.2
✓ Voice matches channel personality (energetic, actionable)
✓ Visual style on-brand (blue + gold, fast cuts)
~ Copy slightly longer than optimal (78 words vs 50-70 target)
✓ Hashtags relevant and niche-specific
Engagement potential: 6.8
✓ Hook strong (first 2 seconds would stop scroll)
✗ Payoff weak ("subscribe" instead of concrete result)
✗ No surprise or emotional arc
~ Call-to-action buried (should be at 3-second mark)
IMPROVEMENT: Add a before/after visual contrast to boost perceived value +40%
OVERALL SCORE: 7.5 (publishable, but needs hook tightening)
POST 2: Tutorial teaser
Brand consistency: 9.1
✓ Tone, visuals, positioning all aligned
✓ Clear value statement in first line
✓ Hashtags researched (trending in niche)
✓ Text hierarchy clear
Engagement potential: 8.4
✓ Strong opening hook ("Without this trick...")
✓ Specificity high (exact tool names, actionable)
✓ Likely to drive profile clicks
✗ Could benefit from controversy or novelty angle
Potential add: "Nobody talks about step 3" to create curiosity
OVERALL SCORE: 8.7 (high-quality, ready to publish)
POST 3: Behind-the-scenes photo
Brand consistency: 7.1
✓ Authentic, personal touch aligns with creator brand
✗ Background slightly cluttered (distracts from subject)
✗ Lighting flat (gold accent missing visual pop)
~ Caption uses casual voice but meanders (145 words, consolidate to 80)
Engagement potential: 5.2
✗ No clear ask or hook (feels like a diary post)
✗ Behind-the-scenes without insight (viewers want to learn, not just see)
✗ Low shareability (niche appeal only)
SUGGESTION: Reframe as "3 things I learned shooting this" to drive engagement +60%
OVERALL SCORE: 6.2 (needs repositioning for traction)
DEVIL'S ADVOCATE ANALYSIS:
"Post 1 looks energetic but sacrifices clarity. Post 2 plays it safe - high polish but lacks edge. Post 3 feels self-focused; viewers want insight, not studio tours."
Counterpoint: Your audience values consistency and trust over viral tricks.
Verdict: Posts 1 & 3 need minor fixes. Post 2 is strong → publish first.
ACTION PRIORITY:
1. Tighten Post 1 hook (add before/after visual)
2. Publish Post 2 as-is
3. Reframe Post 3 or swap for stronger behind-the-scenes angle
CROSS-SESSION SCORE HISTORY:
Session 1: avg 6.8 | Session 2: avg 7.2 | Session 3: avg 7.1 (plateau)
Recommendation: Experiment with new format or storytelling structureAbout this skill
name: self-eval description: Use when Honestly evaluate AI work quality using a two-axis scoring system.
Self Eval
Honestly evaluate AI work quality using a two-axis scoring system. Use after completing a task, code review, or work session to get an unbiased assessment. Detects score inflation, forces devil's advocate reasoning, and persists scores across sessions.
What you get
- Public GitHub repo (alirezarezvani/claude-skills)
- the self-eval skill folder with SKILL.md. Part of a 337-skill / 30-agent / 70-command install.
Customize your output
- Fork the repo and adapt the skill's instructions and references to your workflow.
Example output
Activates automatically when your request matches Self Eval; chains with the other skills, agents, and commands in the collection.
Best for
Creators, builders, and teams using Claude Code.
SKILL.md preview
SKILL.md
---
name: self-eval
description: Use this skill after completing a task, code review, or work session to get an honest, unbiased two-axis assessment of the work's quality.
version: 1.0.0
category: Development / Engineering
author: AgentVolt
license: proprietary
tags:
- development
- engineering
---
# Self Eval
Honestly evaluates completed AI work using a two-axis scoring system, actively resisting the tendency to inflate scores on your own output.
## How to work through it
… (sign up to view the full skill)Sign up to view, copy, and install the full skill
More development skills
View all Development skills →Generate
Scaffold code, tests, and boilerplate from a short spec.
Development
Performance Profiler
Systematic performance profiling for Node.js, Python, and Go applications.
Development
Threat Detection
Use when hunting for threats in an environment, analyzing IOCs, or detecting behavioral anomalies in telemetry.
Development
Agent Harness
Turns a domain folder of skills into a bounded agentic loop: compile a goal into a task plan, execute with the domain's tools, verify every task, retry within a budget.
Development