@mrscraper/opencode 0.3.1 → 0.4.0
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/CHANGELOG.md +17 -0
- package/README.md +3 -3
- package/package.json +3 -3
- package/skills/mrscraper/SKILL.md +5 -5
- package/skills/mrscraper-serp/SKILL.md +25 -54
package/CHANGELOG.md
CHANGED
|
@@ -3,6 +3,23 @@
|
|
|
3
3
|
All notable changes to the MrScraper OpenCode plugin are documented here. The
|
|
4
4
|
project follows [Semantic Versioning](https://semver.org/).
|
|
5
5
|
|
|
6
|
+
## 0.4.0 - 2026-11-02
|
|
7
|
+
|
|
8
|
+
- Updated the `mrscraper-serp` skill for MrScraper MCP server 0.3.0, which
|
|
9
|
+
moved `webSearch` to MrScraper's new web search endpoint. Results are a
|
|
10
|
+
ranked list (rank, title, url, description, source); `region` sets the
|
|
11
|
+
results' country and the new `num_results` asks for 10 to 100 results. The
|
|
12
|
+
`raw`, `format`, `language`, `page`, and `render_js` options are gone, and
|
|
13
|
+
so is the search results URL example. Skills are identical to the MrScraper
|
|
14
|
+
Claude plugin 0.3.0.
|
|
15
|
+
|
|
16
|
+
## 0.3.2 - 2026-10-29
|
|
17
|
+
|
|
18
|
+
- Reworded search as web search in the package
|
|
19
|
+
description, keywords, README, and skills.
|
|
20
|
+
The `webSearch` tool is unchanged. The skills now differ from the MrScraper
|
|
21
|
+
Claude plugin 0.2.2 only in this wording.
|
|
22
|
+
|
|
6
23
|
## 0.3.1 - 2026-10-22
|
|
7
24
|
|
|
8
25
|
- Updated the `mrscraper` skill for the hosted server's three scopes:
|
package/README.md
CHANGED
|
@@ -1,8 +1,8 @@
|
|
|
1
1
|
# MrScraper for OpenCode
|
|
2
2
|
|
|
3
3
|
MrScraper connects [OpenCode](https://opencode.ai) to a hosted MCP server for
|
|
4
|
-
retrieving public web pages as HTML, Markdown, or screenshots,
|
|
5
|
-
|
|
4
|
+
retrieving public web pages as HTML, Markdown, or screenshots, web
|
|
5
|
+
search, and account usage. The plugin also includes three focused skills
|
|
6
6
|
that help OpenCode choose the right workflow and preserve raw page content.
|
|
7
7
|
|
|
8
8
|
The same package works in OpenCode 2 and OpenCode 1.
|
|
@@ -152,7 +152,7 @@ OpenCode 2 reaches MCP tools through Code Mode by default.
|
|
|
152
152
|
| Tool | Purpose |
|
|
153
153
|
| --- | --- |
|
|
154
154
|
| `scraperAPI` | Retrieve a known page as HTML, Markdown, a screenshot, or a combination. |
|
|
155
|
-
| `webSearch` | Discover public pages through
|
|
155
|
+
| `webSearch` | Discover public pages through web search. |
|
|
156
156
|
| `status` | Inspect subscription status, token usage, and remaining tokens. |
|
|
157
157
|
|
|
158
158
|
## Retrieve-first routing
|
package/package.json
CHANGED
|
@@ -1,7 +1,7 @@
|
|
|
1
1
|
{
|
|
2
2
|
"name": "@mrscraper/opencode",
|
|
3
|
-
"version": "0.
|
|
4
|
-
"description": "MrScraper for OpenCode: the hosted MrScraper MCP server plus fetch-first skills for web scraping, structured data extraction, and
|
|
3
|
+
"version": "0.4.0",
|
|
4
|
+
"description": "MrScraper for OpenCode: the hosted MrScraper MCP server plus fetch-first skills for web scraping, structured data extraction, and web search.",
|
|
5
5
|
"type": "module",
|
|
6
6
|
"main": "index.ts",
|
|
7
7
|
"exports": {
|
|
@@ -19,7 +19,7 @@
|
|
|
19
19
|
"web-scraping",
|
|
20
20
|
"data-extraction",
|
|
21
21
|
"scraping",
|
|
22
|
-
"
|
|
22
|
+
"web-search",
|
|
23
23
|
"serp",
|
|
24
24
|
"mcp",
|
|
25
25
|
"research"
|
|
@@ -1,14 +1,14 @@
|
|
|
1
1
|
---
|
|
2
2
|
name: mrscraper
|
|
3
3
|
description: |
|
|
4
|
-
Connect, authenticate, route, and troubleshoot MrScraper MCP, and check account usage with the status tool. For known public URLs, retrieve and preserve page content with mrscraper-fetch first; use mrscraper-serp for query-first discovery through
|
|
4
|
+
Connect, authenticate, route, and troubleshoot MrScraper MCP, and check account usage with the status tool. For known public URLs, retrieve and preserve page content with mrscraper-fetch first; use mrscraper-serp for query-first discovery through web search.
|
|
5
5
|
---
|
|
6
6
|
|
|
7
7
|
# MrScraper MCP
|
|
8
8
|
|
|
9
9
|
Use this skill for connection and authentication checks, tool selection, shared
|
|
10
10
|
response handling, account status, and troubleshooting. Load the focused fetch
|
|
11
|
-
or SERP skill for page retrieval and
|
|
11
|
+
or SERP skill for page retrieval and web search.
|
|
12
12
|
|
|
13
13
|
## Step 1 — Verify or Connect MrScraper MCP
|
|
14
14
|
|
|
@@ -65,7 +65,7 @@ complete.
|
|
|
65
65
|
| User outcome | MCP tool or skill |
|
|
66
66
|
| --- | --- |
|
|
67
67
|
| Retrieve and preserve a known public page as HTML, Markdown, or a screenshot | [mrscraper-fetch](../mrscraper-fetch/SKILL.md) with scraperAPI |
|
|
68
|
-
| Discover relevant pages from a
|
|
68
|
+
| Discover relevant pages from a web search query | [mrscraper-serp](../mrscraper-serp/SKILL.md) with webSearch, then retrieve selected pages |
|
|
69
69
|
| Check the plan, token usage, or remaining tokens | status |
|
|
70
70
|
|
|
71
71
|
### Retrieve first, extract locally
|
|
@@ -158,8 +158,8 @@ account usage, not the progress of a request.
|
|
|
158
158
|
replaced by scraperAPI, webSearch, and status.
|
|
159
159
|
- Page incomplete or missing dynamic content: load mrscraper-fetch and revise
|
|
160
160
|
page-loading inputs deliberately.
|
|
161
|
-
- Search results weak: load mrscraper-serp and refine the query,
|
|
162
|
-
|
|
161
|
+
- Search results weak: load mrscraper-serp and refine the query, country, or
|
|
162
|
+
number of results.
|
|
163
163
|
|
|
164
164
|
## Limits
|
|
165
165
|
|
|
@@ -1,7 +1,7 @@
|
|
|
1
1
|
---
|
|
2
2
|
name: mrscraper-serp
|
|
3
3
|
description: |
|
|
4
|
-
Discover public pages through
|
|
4
|
+
Discover public pages through web search with MrScraper's webSearch tool. Use when no target URL is known or the user asks to inspect web search results. After discovery, retrieve selected pages with mrscraper-fetch.
|
|
5
5
|
---
|
|
6
6
|
|
|
7
7
|
# Discover Pages with MrScraper MCP Web Search
|
|
@@ -13,86 +13,58 @@ status, or broader routing.
|
|
|
13
13
|
|
|
14
14
|
## Step 1 — Build the Query
|
|
15
15
|
|
|
16
|
-
Include the important entity, attribute, location, or
|
|
17
|
-
|
|
16
|
+
Pass the query in url. Include the important entity, attribute, location, or
|
|
17
|
+
date:
|
|
18
18
|
|
|
19
19
|
{
|
|
20
|
-
"url": "web scraping best practices"
|
|
21
|
-
"raw": false
|
|
20
|
+
"url": "web scraping best practices"
|
|
22
21
|
}
|
|
23
22
|
|
|
24
|
-
The input can also be a
|
|
23
|
+
The input can also be a search results URL; its q parameter becomes the query.
|
|
25
24
|
|
|
26
|
-
|
|
27
|
-
"url": "https://www.google.com/search?q=web+scraping&gl=us&hl=en&start=20",
|
|
28
|
-
"raw": false
|
|
29
|
-
}
|
|
30
|
-
|
|
31
|
-
For a Google URL, the tool reads q, gl, hl, and start. Explicit region,
|
|
32
|
-
language, and page inputs take precedence.
|
|
33
|
-
|
|
34
|
-
## Step 2 — Set Locale and Pagination
|
|
25
|
+
## Step 2 — Set the Country and Result Count
|
|
35
26
|
|
|
36
|
-
Set
|
|
27
|
+
Set region to a two-letter country code when results must come from one
|
|
28
|
+
country. It defaults to us:
|
|
37
29
|
|
|
38
30
|
{
|
|
39
31
|
"url": "running shoes",
|
|
40
|
-
"
|
|
41
|
-
"region": "id",
|
|
42
|
-
"language": "id",
|
|
43
|
-
"page": 2
|
|
44
|
-
}
|
|
45
|
-
|
|
46
|
-
Start with the first page unless the user requests broader coverage. Inspect
|
|
47
|
-
the current page before requesting another.
|
|
48
|
-
|
|
49
|
-
## Step 3 — Choose the Output
|
|
50
|
-
|
|
51
|
-
raw defaults to true, which returns result-page HTML as the Python SDK does.
|
|
52
|
-
Set raw=false for parsed JSON; organic results are in data.data.organic, next
|
|
53
|
-
to sections such as people_also_ask and pagination.
|
|
54
|
-
|
|
55
|
-
Request result-page HTML when the page itself is needed:
|
|
56
|
-
|
|
57
|
-
{
|
|
58
|
-
"url": "iphone 17"
|
|
32
|
+
"region": "id"
|
|
59
33
|
}
|
|
60
34
|
|
|
61
|
-
|
|
35
|
+
The search returns 10 results by default. Ask for more, up to 100, only when
|
|
36
|
+
the user needs broader coverage:
|
|
62
37
|
|
|
63
38
|
{
|
|
64
|
-
"url": "
|
|
65
|
-
"
|
|
66
|
-
"
|
|
39
|
+
"url": "running shoes",
|
|
40
|
+
"region": "id",
|
|
41
|
+
"num_results": 30
|
|
67
42
|
}
|
|
68
43
|
|
|
69
44
|
### Parameters
|
|
70
45
|
|
|
71
46
|
| MCP input | Default | Request mapping | Use |
|
|
72
47
|
| --- | --- | --- | --- |
|
|
73
|
-
| url | required | Body query | Search query or
|
|
74
|
-
|
|
|
75
|
-
|
|
|
76
|
-
| region | URL value or omitted | Body region | Google result country, such as us or id. |
|
|
77
|
-
| language | URL value or omitted | Body language | Google result language, such as en or id. |
|
|
78
|
-
| page | URL value or omitted | Body page | Positive, one-based results page. |
|
|
79
|
-
| render_js | false | Body renderJs | Render JavaScript-dependent result features. |
|
|
48
|
+
| url | required | Body query | Search query, or a search results URL whose q value is used. |
|
|
49
|
+
| region | us | Body country | Two-letter result country, such as us or id. |
|
|
50
|
+
| num_results | 10 | Body numResults | Number of results, from 10 to 100; fewer may come back. |
|
|
80
51
|
| timeout | 600 | MCP upstream deadline | Maximum wait in seconds; not sent in the search body. |
|
|
81
52
|
|
|
82
53
|
Authentication is handled by the MCP client's OAuth 2.1 connection and is
|
|
83
54
|
never a tool input.
|
|
84
55
|
|
|
85
|
-
## Step
|
|
56
|
+
## Step 3 — Inspect the Results
|
|
86
57
|
|
|
87
58
|
The complete response envelope is available in MCP structuredContent and as
|
|
88
59
|
formatted JSON text. Prefer structuredContent. Check isError, error, and
|
|
89
|
-
status_code before using data.
|
|
60
|
+
status_code before using data. The results are a ranked list in data.data;
|
|
61
|
+
each result has rank, title, url, description, and source.
|
|
90
62
|
|
|
91
63
|
When the user requests a search artifact, save the successful data or full
|
|
92
64
|
envelope with local file-writing capability; webSearch has no output-path
|
|
93
65
|
input.
|
|
94
66
|
|
|
95
|
-
## Step
|
|
67
|
+
## Step 4 — Continue with Relevant URLs
|
|
96
68
|
|
|
97
69
|
Select only URLs relevant to the user's goal, then load
|
|
98
70
|
[mrscraper-fetch](../mrscraper-fetch/SKILL.md) to retrieve and inspect each
|
|
@@ -102,15 +74,14 @@ Run independent follow-up URLs in parallel when the environment supports safe
|
|
|
102
74
|
parallel execution. Keep the number of pages proportional to the requested
|
|
103
75
|
coverage. If the user only requested URL discovery, stop after the search.
|
|
104
76
|
|
|
105
|
-
## Step
|
|
77
|
+
## Step 5 — Handle Weak or Missing Results
|
|
106
78
|
|
|
107
|
-
- Set
|
|
79
|
+
- Set region when the results come from the wrong country.
|
|
108
80
|
- Rewrite noisy queries around the important entity, attribute, and location.
|
|
109
|
-
-
|
|
110
|
-
- Retry once with render_js=true when a dynamic result feature is required.
|
|
81
|
+
- Raise num_results only after checking the first results.
|
|
111
82
|
- Increase timeout when the MCP request deadline is too short.
|
|
112
83
|
- Use [mrscraper](../mrscraper/SKILL.md) when the connection or authentication
|
|
113
84
|
fails.
|
|
114
85
|
|
|
115
|
-
webSearch is for public
|
|
86
|
+
webSearch is for public web search, not for discovering URLs within one
|
|
116
87
|
known site.
|