Weekly Improve Loop: Small Tests, Kept Winners
Automate weekly sales optimization with data-driven KPI analysis, single-variable experiments, and ledger logging.
$12
Works with the AI Agents you already use
Claude Code
Cursor
Codex CLI
Muse
Dots+22 moreClaude CodeCursorCodex CLIMuseDots
OpenClaw
Manus
Grok
Grok Bot+18 agentsWeekly Improve Loop: Small Tests, Kept Winners
Example session with this skill installed
Run the weekly review for 'gadget-sleeve' for the week of 2026-10-12. Check EXP-05 and see if we can improve the click-through rate on the primary listing.
- Read your context and instructions
- Compiled the weekly improve loop
- Generated the data export
Weekly review completed for gadget-sleeve. I have updated experiments.csv and logged a new learning regarding title performance.
WEEKLY REVIEW - gadget-sleeve - week of 2026-10-12
KPIS VS TARGET: listing visits 88 / 120 behind | post views 410 / 400 on | orders 2 / 3 behind | conversion 2.2% / 2.5% behind
**EXPERIMENTS:** EXP-05 title keywords: A 42 visits vs B 46 visits (+9%, both below 100 min sample) -> extend
**NEW EXPERIMENT:** EXP-06 image order on L2, A=IMG-01 (main) vs B=IMG-02 (lifestyle), KPI listing visits, min 100 visits, ends 2026-10-26
**TWEAKS APPLIED:** Swapped tags on L3 to approved "minimalist-tech" set; updated post schedule for Tuesday to 09:00 per tweak line.
LEARNINGS: LRN-014 No significant difference found between cost-first and feature-first titles at current traffic levels (EXP-05).
NOTE: Order volume remains low (2 orders); data is insufficient to suggest offer changes.
PROPOSALS (need your OK):
- Reduce weekly listing visit target from 120 to 100 to reflect actual baseline traffic.
- Approve new variant C12 (gift-focus) for future testing.
DECISION: Should we extend EXP-05 for another 7 days or drop it to test a new image variant instead?
Next steps
- Review the proposal to lower the listing visit target.
- Approve or reject content variant C12 in
content.md. - Provide a decision on extending EXP-05.
experiments.csv
CSV · data export
Example file from a real run - the skill writes it into your workspace.
Connects securely to your tools. The creator never sees your data.
About this skill
Every week it reads your sales ledger, runs at most two one-variable tests, keeps what wins by a clear margin, writes down what it learned and makes small reversible tweaks inside the plan you approved.
You give it: the sales-system ledger after the week's numbers are in. You get back: a WEEKLY REVIEW: experiments started, kept or dropped (with the numbers), learnings with evidence, tweaks applied, and proposals (new copy, new lines, KPI target changes) as cards for you. It stops at: tweaks that use approved content and stay inside plan limits, or pausing something. Raising spend, frequency or targets, new copy and new channels all go to you. It doesn't declare a winner on thin data.
Includes a worked example and 3 test scenarios. Works with Grok Bot and other SKILL.md agents. Not affiliated with xAI.
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 passes our trust checks below.
Security scanned
Passed 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 scanned
Verified clean 3 days ago
- Passed all security checks, Safe to install