mcpField
K
kalodata

kalodata

github.com/sailtonightMarketing & growth188 installsUpdated 2d ago
Repository

About kalodata

kalodata is a skill for TikTok Shop e-commerce research, with a bundled Python CLI for querying products, shops, creators, videos, livestreams, and categories. Results are available in TOON or JSON, with ranking, detail, and analysis workflows.

Key features

  1. 01Query product, shop, and creator rankings
  2. 02Explore video, livestream, and category data
  3. 03Retrieve details, comments, and product specs
  4. 04Output TOON or JSON
  5. 05Check API credit balance and logs

Use cases

  1. 01Research trending TikTok Shop products
  2. 02Find and evaluate creators
  3. 03Analyze category revenue and shop performance
  4. 04Review shoppable video and livestream metrics

Install

Copy the command and run it in a terminal.

$npx skills add https://github.com/sailtonight/kalodata-skill --skill kalodata

After installing, type/kalodatain Claude Code to confirm it loaded.

Bundled files

References and scripts installed along with SKILL.md.

references/
playbooks.md
scripts/
kalo.py
scripts/kalodata/
__init__.py
__main__.py
api.py
cli.py
config.py
core.py
errors.py
mcp_server.py
onboarding.py
reference.py
render.py
toon.py
update.py
validate.py
scripts/kalodata/commands/
__init__.py
category.py
common.py
creator.py
credit.py
livestream.py
meta.py
product.py
shop.py
video.py

Details

From the project README

KaloData TikTok Shop analytics (kalo)

TikTok Shop e-commerce analytics from KaloData — rankings and details for products, shops, creators, videos, livestreams and categories.

Run

The CLI ships inside this skill (scripts/kalo.py, Python 3.11+ stdlib only — nothing to install):

python3 <this-skill-dir>/scripts/kalo.py product rank --region US

All examples below write kalo as shorthand for python3 <this-skill-dir>/scripts/kalo.py.

Output is TOON (token-oriented) on stdout; add --json to any command for raw JSON. Errors are structured on stdout with actionable help: lines. Exit codes: 0 ok, 1 error, 2 usage.

Setup — check this before the first query

Every command needs a KaloData API key (sent as the secret-key header). Before the first query of a session run kalo config and look at api_key. If it says missing, stop — don't guess a key and don't retry the query. Walk the user through these three steps and wait:

  1. Get a key — sign in at https://www.kalodata.com/open-center/account and click generate key. No account yet? Register at https://www.kalodata.com/open-center/home first. If they already call the KaloData API on credit-based billing, that same key works here — no new key needed.

  2. Save it — kalo config set --key <API_KEY> (stored in ~/.config/kalodata/config.toml, mode 0600), or set KALODATA_API_KEY in the environment.

  3. Credits — every query spends Kalo Credits. Measured costs per call:

    callcredits
    any rank or detail lookup0.2
    product images0.01
    creator images0.1
    product specs0.1
    video url0.5
    product comments (only when the upstream cache refreshes)1
    credit balance, credit logsfree

    Costs were measured against production; rank/detail lookups bill 0.2 flat under this skill's source-type: SKILL header. A playbook fans out into many calls, so it lands at 1–2 credits. kalo credit shows the remaining balance; top up at https://www.kalodata.com/pricing.

Don't announce or ask about credit costs for queries the user asked for — just run them. Bring up cost only when the balance may not cover a big fan-out (a data-gathering playbook, dozens of detail lookups: check kalo credit first), or when the user asks.

Commands

commandwhat
kalo product ranktop products; --category --keyword --price --launch --shop --creator
kalo product detail <id...>price range, shop id, revenue trend, product description
kalo product comments <id...>review insight: pain points, positives, usage scenarios (1 credit per refresh)
kalo product specs <id...>selling points (highlights) and spec attributes; --lang zh-CN translates
kalo product images <id...>product gallery image URLs
kalo shop ranktop shops; --type BRAND|RETAILER --keyword
kalo shop detail <id...>revenue split, top product ids
kalo creator ranktop creators; --followers --engagement --product
kalo creator detail <id|@handle...>contact email/handle, GPM, trend; handle match is fuzzy — verify the returned creator_handle
kalo creator images <id...>avatar URLs, up to 100 creator ids
kalo video ranktop shoppable videos; --product --creator --keyword --ai
kalo video detail <id...>full video metrics + trend
kalo video url <id...>playable mp4 URLs
kalo live ranktop livestreams; --keyword
kalo live detail <id...>viewers, GPM, top product ids
kalo category ranktop categories; --level 1|2|3
kalo category detail <id...>category revenue split + trend
kalo category search <kw>resolve category ids by keyword
kalo creditremaining credit balance for this key
kalo credit logswhat credits were spent on; --range --type --page
kalo configshow/set credentials and defaults

Conventions

  • All list commands take --region --range --page (1-5) --limit (5-100) --sort --asc --fields --json
  • Regions: US GB ID TH VN PH MY SG JP MX DE IT FR ES BR; ranges like last7Day, last30Day, yyyy-MM, or yyyy-MM-dd~yyyy-MM-dd
  • Numeric ranges are "min-max" with both ends numeric, e.g. --revenue 1000-50000
  • Add --images on product/shop rank or product detail to include image URLs; creator avatars come from kalo creator images (creator rank has no image column)
  • Signed media URLs (product images, creator images, video url) expire about 5 minutes after the call — hand them to the user or download them now, never cache them
  • Run kalo <command> --help for per-command flags and examples

Typical flows

  • Trending research: kalo product rank --region US --range last7Day → kalo product detail <id> → kalo video rank --product <id>
  • Creator scouting: kalo creator rank --category <id> --followers 10000-1000000 → kalo creator detail <id> (has contact email)
  • Category drilldown: kalo category search "beauty" → kalo category detail <id> → kalo shop rank --category <id>

Playbooks

For complex analysis — product/creator diagnosis, category sourcing research, creator recommendation, viral-video methodology, shot-by-shot script writing — read references/playbooks.md first and follow its mandatory steps (always benchmark against same-category / same-price-band / same-follower-tier peers before drawing conclusions).

Comments

Markdown supported