Analytics review
Claude Code pack for Instagram & Facebook content creators: FTC/Meta disclosure, copyright, brand-safety & no-fabricated-stats guardrails + skills for Reels/carousels/Stories/captions/repurposing/sponsorships, loops & review agents. Not legal advice.
npx -y skills add m-binimran/creator-pack --skill analytics-reviewAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
One thing to look at
- 3 stars3 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
What its author says it does
Copied from the file, not written here
Interpret Instagram/Facebook Insights and turn them into next actions - what's working, what to do more of, what to cut. Use when reviewing performance from supplied metrics.
SKILL.md
1.4 KB, as published. Nobody here has run it
analytics-review
Numbers into decisions. Read the right metrics for the goal, then act.
Read the right metric per goal
- Reach/growth: reach, follows from a post, shares, watch-through on Reels.
- Resonance/value: saves and shares (strongest value signals on Meta), comments.
- Retention: Reel average watch time / replays; carousel swipe-through; Story completion + tap-backs.
- Action: profile visits, link taps, DMs.
Process
- Use the creator's real numbers (supplied or pulled via the Meta connector) - don't invent benchmarks.
- Find patterns: which hooks/formats/topics over-index on saves/shares/retention? Where do people drop?
- Recommend: do more of what works (format/hook/topic), fix the drop-off point, stop what underperforms.
- One experiment for next week (e.g. test a new hook style on Reels).
Output
- A short read: top performers + why, the drop-off insight, 3 concrete next actions, 1 experiment. Cite the
actual numbers; mark anything missing
[verify].
Guardrails
- No fabricated benchmarks or made-up metrics; work from real data (
truth-telling). - Don't over-read tiny samples; note when data is thin.