Registry skill by asgard-ai-platform
Taiwan E-Commerce Benchmarks asgard-ai-platform/tw-ecom-analytics-benchmarks
Taiwan e-commerce benchmark ranges for CVR, ROAS, LTV, AOV, repeat rate, cart-abandon — segmented by vertical (3C, 美妝, 服飾, 母嬰, 生鮮) and channel (DTC, Shopee, momo).
What it changes View full report → Passed 18 of 22 test cases with the skill, and 18 of 22 without it.
Without skill 82%
With skill 82%
Checks it passed that the model alone missed 2 fixed16 passed either way2 still failing2 broke
Pass-rate lift 0 pp
Tokens +21%
Agent turns 0%
Response time +13% Model gemini-3.6-flashMeasured 8/28/2026 v1 18 of 22 cases passed with skill
Use this skill Claude Code DecimalAI agent CLI & files
Save it into your project's skills folder. Claude Code picks it up in its next session.
curl -fsSL --create-dirs -o .claude/skills/asgard-ai-platform-tw-ecom-analytics-benchmarks/SKILL.md https://app.decimal.ai/s/asgard-ai-platform-tw-ecom-analytics-benchmarks/SKILL.md Copy command Copy SKILL.md
Want it pinned and updatable? pip install "decimalai>=0.15", then decimalai skills add asgard-ai-platform-tw-ecom-analytics-benchmarks, and start Claude Code with claude --plugin-dir "$(git rev-parse --show-toplevel)/.agents"
Linked stays up to date: your agents get the author's new versions automatically.
Editable copy is yours to change; the author's updates won't reach it.
Install linked skill Create editable copy
1 · One-time — install the CLI (
docs ↗ )
$ pip install "decimalai>=0.15"Copy
2 · Get the skill — writes .agents/skills/asgard-ai-platform-tw-ecom-analytics-benchmarks/ and pins it in .decimal/skills.lock
$ decimalai skills add asgard-ai-platform-tw-ecom-analytics-benchmarksCopy
Copy link Copy README badge Copy SKILL.md No CLI? Get the same file — curl · raw file $ curl -fsSL --create-dirs -o .claude/skills/asgard-ai-platform-tw-ecom-analytics-benchmarks/SKILL.md https://app.decimal.ai/s/asgard-ai-platform-tw-ecom-analytics-benchmarks/SKILL.mdCopy
Safety checks: 3 of 3 passed
Code scan Clean Intent Passed Content Clean
Overview Benchmark Contents Activity Versions 1 Ratings 0
See it in action One of 2 test cases the skill fixed. Same model, same prompt, with and without the skill.
Prompt How do I change the reporting attribution model from Data-Driven to Last Click inside the Google Analytics 4 Admin panel for our Taiwan web property?
What was checked, without and with the skill Checked Without skill With skill Outlines the step-by-step navigation in GA4 Admin under Property Settings > Data Settings > Attribution Settings. Does not analyze channel ROAS or benchmark conversion rates.
Show both answers
What's inside The instructions your model reads when the skill loads.
SKILL.md 46 lines single file
Taiwan E-Commerce Benchmarks
STATUS: SKELETON — body pending.
When to use this skill
A merchant asks "is my CVR / ROAS / LTV good?"
Sizing a business case (revenue / ad-spend projections)
Comparing performance across verticals or channels
Investor / banker deck benchmarks
Do NOT use when
Instrumentation itself → tw-ecom-analytics-ga4
Unit economics framework → biz-unit-economics
Core concepts
TODO: benchmark tables by vertical × channel, sourced ranges with citations.
Decision tree
TODO: merchant profile → applicable benchmark row.
Implementation guidance
TODO: comparison template, outlier-flag criteria, what to do when outside range.
Gotchas
TODO: 5-6 pitfalls (stale benchmarks, vertical misclassification, channel-mix distortion, attribution-model divergence, peak-period distortion).
IRON LAW
TODO (candidate: "All benchmark ranges must cite source + year. An uncited number is worse than no number.").
Output Format
TODO.
When it activates Automatically, when a task matches.
Use when a TW merchant asks 'is my CVR / ROAS good?' or when sizing a business case. Source discipline: cite industry report or vendor data; mark undocumented ranges as estimates. STATUS: SKELETON — body pending.
Track record Benchmark 18 / 22
test cases passed, vs 18 / 22 without
Adoption
Users 0
Times used 0 View activity → Safety and source details
SkillSafety ✓ 3 of 3 passed ✓ Code scan 15 of 15 checks clean
0 critical · 0 warning · 0 info — scanner v4 — 8/3/2026
View 15 scanner checks live_secret remote_code_execution reverse_shell data_exfiltration ssrf_metadata agent_snooping anti_refusal destructive_commands obfuscated_unicode unicode_homoglyph opaque_blob suspicious_url tool_overreach trigger_abuse prompt_injection_phrasing
✓ Intent Safe
LLM judge · does the skill do what it claims
✓ Content Clean
LLM judge · policy & harmful-content review
Source Imported asgard-ai-platform / skills
tw-ecom-analytics-benchmarks/SKILL.md
View on GitHub ↗ License MITImported 8/3/2026Version v1