Works with the AI Agents you already use
Claude Code
Cursor
Codex CLI
Muse
Dots+21 moreClaude CodeCursorCodex CLIMuseDots
Manus
Grok
Grok Bot+18 agentsReel Performance Diagnosis Skill
Diagnose Reel performance through retention curves, engagement ratios, and structural audits to find why videos stall.
$5.99
Reel Performance Diagnosis Skill
Example session with this skill installed
Analyze this Reel: 60s duration, 10k views, 8k reach, 45% completion, 200 likes, 5 shares, 80 saves. The hook was '3 tools I use for SEO', but retention dropped 30% at the 15-second mark.
- Read your context and instructions
- Compiled the reel performance diagnosis
The Reel shows strong retention and saveability but very low shareability, which likely restricted its viral reach. The significant drop at 15 seconds suggests a failure to transition from the hook to the value delivery.
Reel Performance Diagnosis
KPI Diagnosis
| Metric | Reel | Benchmark (Est) | Difference | Interpretation |
|---|---|---|---|---|
| Reach | 8,000 | 12,000 | -33% | Lower than average distribution |
| Completion | 45% | 25% | +20% | Exceptional mid-to-late retention |
| Save Rate | 1.0% | 0.2% | +5x | High utility/value perception |
| Share Rate | 0.06% | 0.5% | -88% | Minimal social currency/virality |
Retention Diagnosis
- Opening: Strong initial hook; 8,000 reach led to 10,000 plays (some rewatching).
- Biggest drop-off: 30% loss at 0:15.
- Mid-video behavior: Extremely stable after the 15s mark.
Likely structural cause: The transition from the "3 tools" promise to the first tool was too slow or lacked a visual pattern interrupt.
- Confidence: High (Data-backed).
Content Structure Audit
| Section | Observation | Impact | Recommended Change |
|---|---|---|---|
| Hook | '3 tools I use for SEO' | High initial interest | Keep the hook but speed up the 'why'. |
| Setup | Transition at 15s | 30% viewer loss | Move first tool reveal to 0:05. |
| Value | SEO tool walkthroughs | High saves (80) | Maintain this level of detail. |
| CTA | Not provided | Unknown | Add a 'Save for later' prompt. |
Root-Cause Diagnosis
- Finding: High Save-to-Share Ratio.
- Evidence: 80 saves vs 5 shares.
- Classification: Confirmed by data.
Likely impact: The content is "selfish" value (useful to the individual) rather than "social" value (useful to tell others about). This limits the reach multiplier.
Priority Fixes
- Accelerate the Setup: Reduce the intro time between the hook and the first tool reveal to under 5 seconds.
Visual Pattern Interrupt: Add a text overlay or b-roll cut exactly at the 12-second mark to bridge the 15-second drop-off.
3. Shareable Hook: Experiment with a "POV" or "Common Mistake" hook to increase shareability.
Next-Reel Test Plan
- Hypothesis: Moving the first value point to 0:05 will reduce the 15s drop-off by 50%.
- Variable: Pacing of the first transition.
- Control: Same topic (SEO tools) and same visual style.
- Success metric: Retention percentage at the 15-second mark.
Next steps
- Review the raw footage at 0:12-0:15 to identify dead air or repetitive phrasing.
- Re-edit the same script with a faster transition and repost to test the pacing hypothesis.
- Add a specific prompt at the end asking users to share with a teammate.
Connects securely to your tools. The creator never sees your data.
About this skill
The problem
High-quality Reels often fail to gain traction, but basic analytics don't explain why. Most creators can see that a video underperformed but cannot pinpoint whether the issue was the hook, the mid-video pacing, or a mismatch between the promise and the payoff.
What it does
- Identifies specific retention drop-off points and maps them to script or visual transitions.
- Separates distribution signals from content quality to determine if reach was limited by the algorithm or the audience.
- Audits content structure, including hooks, pattern interrupts, and CTA friction, based on performance facts.
- Classifies findings into evidence-backed root causes versus plausible hypotheses.
- Generates controlled A/B test plans for the next Reel to isolate variables and improve results.
Why this beats prompting it yourself
General LLMs tend to offer vague advice like "make it more engaging" or "use trending audio." This skill enforces a data-first workflow that treats metrics as evidence, preventing the hallucination of algorithmic causes and ensuring every recommendation is tied to a specific observation in your retention curve or engagement ratio.
Use cases
- Post-mortem analysis for a viral video that failed to convert viewers into followers.
- Diagnosing why a high-retention Reel received unusually low reach.
- Comparing a successful Reel against a recent flop to identify the winning variable.
- Optimizing script structures for creators who have hit a plateau in average watch time.
Known limitations
Cannot verify claims if metrics are missing. Does not account for external platform outages or broad algorithm updates. Requires manual input of analytics data as it does not have direct API access to social accounts.
Cannot diagnose performance from views alone. Requires manual input of analytics. Does not have direct API access to social platforms. Accuracy depends on the granularity of data provided.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before any payment.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scan
Security scan passed 1 month ago
Permissions
No special permissions declared or detected