webmd

Name: webmd
Author: QSong-github

132stars4forks

How to install webmd?

qsong-github/drugclaw/webmd

$npx -y skills add QSong-github/DrugClaw --skill webmd

Installs into the current project.

Summary

Query the WebMD Drug Reviews dataset (~362 k patient reviews, 2007–2020). Use whenever the user asks about patient-reported drug effectiveness, ease of use, satisfaction ratings, side effects, or reviews for a specific drug or medical condition.

Files · 1

View on GitHub

SKILL.md

1	# WebMD Drug Reviews Query Skill
2
3	Search patient reviews by drug name or medical condition. Auto-fallback: if no drug matches, searches conditions.
4
5	\| Input Example \| Match Logic \|
6	\|---\|---\|
7	\| `Lipitor` \| substring on `Drug` column \|
8	\| `High Blood Pressure` \| substring on `Condition` column \|
9	\| any text \| try `Drug` first → fall back to `Condition` \|
10
11	## API
12
13	\| Function \| Input \| Returns \|
14	\|---\|---\|---\|
15	\| `load_reviews(path)` \| CSV path (default `DATA_PATH`) \| list[dict] \|
16	\| `search(rows, entity)` \| single entity string \| list[dict] \|
17	\| `search_by_drug(rows, drug)` \| drug name string \| list[dict] \|
18	\| `search_by_condition(rows, cond)` \| condition string \| list[dict] \|
19	\| `search_batch(rows, entities)` \| list of entity strings \| dict[str, list[dict]] \|
20	\| `summarize(hits, entity)` \| matched rows + label \| compact text (counts, avg ratings, top conditions/drugs) \|
21	\| `to_json(hits)` \| matched rows \| list[dict] \|
22
23	## Usage
24
25	See `if __name__ == "__main__"` block in `10_WebMD_Drug_Reviews.py` for runnable examples covering: single drug lookup, single condition lookup, batch search, and JSON output.
26
27	## Data
28
29	- Source: `webmd.csv` (Kaggle – rohanharode07/webmd-drug-reviews-dataset)
30	- Columns: `Drug`, `Condition`, `Reviews`, `Side`, `Age`, `Sex`, `Effectiveness`, `EaseofUse`, `Satisfaction`
31	- Ratings: 1-5 star scale for Effectiveness, EaseofUse, Satisfaction
32	- Path: `DATA_PATH` variable in `10_WebMD_Drug_Reviews.py`
33	(`resources_metadata/drug_review/WebMDDrugReviews/webmd.csv`)

People who liked this also viewed

6 picks

Type

webmd

132stars4forks

View on GitHub

How to install webmd?

qsong-github/drugclaw/webmd

$npx -y skills add QSong-github/DrugClaw --skill webmd

Installs into the current project.

Summary

Files · 1

View on GitHub

SKILL.md

1	# WebMD Drug Reviews Query Skill
2
3	Search patient reviews by drug name or medical condition. Auto-fallback: if no drug matches, searches conditions.
4
5	\| Input Example \| Match Logic \|
6	\|---\|---\|
7	\| `Lipitor` \| substring on `Drug` column \|
8	\| `High Blood Pressure` \| substring on `Condition` column \|
9	\| any text \| try `Drug` first → fall back to `Condition` \|
10
11	## API
12
13	\| Function \| Input \| Returns \|
14	\|---\|---\|---\|
15	\| `load_reviews(path)` \| CSV path (default `DATA_PATH`) \| list[dict] \|
16	\| `search(rows, entity)` \| single entity string \| list[dict] \|
17	\| `search_by_drug(rows, drug)` \| drug name string \| list[dict] \|
18	\| `search_by_condition(rows, cond)` \| condition string \| list[dict] \|
19	\| `search_batch(rows, entities)` \| list of entity strings \| dict[str, list[dict]] \|
20	\| `summarize(hits, entity)` \| matched rows + label \| compact text (counts, avg ratings, top conditions/drugs) \|
21	\| `to_json(hits)` \| matched rows \| list[dict] \|
22
23	## Usage
24
25	See `if __name__ == "__main__"` block in `10_WebMD_Drug_Reviews.py` for runnable examples covering: single drug lookup, single condition lookup, batch search, and JSON output.
26
27	## Data
28
29	- Source: `webmd.csv` (Kaggle – rohanharode07/webmd-drug-reviews-dataset)
30	- Columns: `Drug`, `Condition`, `Reviews`, `Side`, `Age`, `Sex`, `Effectiveness`, `EaseofUse`, `Satisfaction`
31	- Ratings: 1-5 star scale for Effectiveness, EaseofUse, Satisfaction
32	- Path: `DATA_PATH` variable in `10_WebMD_Drug_Reviews.py`
33	(`resources_metadata/drug_review/WebMDDrugReviews/webmd.csv`)

People who liked this also viewed

6 picks

Type