mcp-scraper 0.58.0 → 0.60.0
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 +18 -0
- package/README.md +1 -1
- package/dist/bin/api-server.cjs +415 -100
- package/dist/bin/api-server.cjs.map +1 -1
- package/dist/bin/api-server.js +1 -1
- package/dist/bin/mcp-scraper-cli.cjs +1 -1
- package/dist/bin/mcp-scraper-cli.cjs.map +1 -1
- package/dist/bin/mcp-scraper-cli.js +1 -1
- package/dist/bin/mcp-scraper-install.cjs +1 -1
- package/dist/bin/mcp-scraper-install.cjs.map +1 -1
- package/dist/bin/mcp-scraper-install.js +1 -1
- package/dist/bin/mcp-stdio-server.cjs +7 -6
- package/dist/bin/mcp-stdio-server.cjs.map +1 -1
- package/dist/bin/mcp-stdio-server.js +2 -2
- package/dist/{chunk-P32KEY3D.js → chunk-AIA3MFT2.js} +8 -7
- package/dist/chunk-AIA3MFT2.js.map +1 -0
- package/dist/chunk-AQA2ZL4X.js +7 -0
- package/dist/chunk-AQA2ZL4X.js.map +1 -0
- package/dist/editorial-reading-room/assets/app.js +17 -3
- package/dist/{server-J4YGBOTS.js → server-IRKNU6LO.js} +395 -96
- package/dist/server-IRKNU6LO.js.map +1 -0
- package/package.json +1 -1
- package/dist/chunk-EJ6KU4PC.js +0 -7
- package/dist/chunk-EJ6KU4PC.js.map +0 -1
- package/dist/chunk-P32KEY3D.js.map +0 -1
- package/dist/server-J4YGBOTS.js.map +0 -1
package/CHANGELOG.md
CHANGED
|
@@ -4,6 +4,24 @@ All notable changes to MCP Scraper are documented here. The format is based on [
|
|
|
4
4
|
|
|
5
5
|
## [Unreleased]
|
|
6
6
|
|
|
7
|
+
## [0.60.0] - 2026-08-17
|
|
8
|
+
|
|
9
|
+
### Added
|
|
10
|
+
|
|
11
|
+
- Transparent Commons has its own semantic search. `commons_index_documents` had carried `embedding_status`, `vector_ref` and provider columns since launch with 231 documents queued and none ever embedded. An embedding worker now drains that queue on every cron tick, using the same Jina model Memory uses, into a platform-scoped `commons_index_vectors` table in the shared Postgres. Nothing about it touches a personal Memory vault: the corpus is public and account-independent.
|
|
12
|
+
- `commons_search_entities` gained `mode`: `hybrid` (default) fuses semantic candidates with exact term matching and ranks title-first, `lexical` matches terms only, `semantic` ignores term matching. Semantic candidates enter the result set rather than only re-ranking lexical hits, so a query with no shared vocabulary still finds the right entity — "tools that let an AI assistant browse the web" returns Web scraping, Model Context Protocol and AI agent, where lexical returns nothing.
|
|
13
|
+
- `POST /commons/index/embed` and `GET /commons/index/status` for operating the index directly.
|
|
14
|
+
|
|
15
|
+
## [0.59.0] - 2026-08-17
|
|
16
|
+
|
|
17
|
+
### Added
|
|
18
|
+
|
|
19
|
+
- Every published editorial article now has its own URL. `GET /commons/publications/{subdomain}/articles/{articleSlug}/site` server-renders a standalone article page with its own canonical, Open Graph tags, and `Article` JSON-LD that declares the edition as its `CollectionPage`. An edition-scoped form is available at `/editions/{editionSlug}/articles/{articleSlug}/site`.
|
|
20
|
+
|
|
21
|
+
### Changed
|
|
22
|
+
|
|
23
|
+
- The reading-room app reads and writes real article paths instead of an `?article=` query parameter, so an article is shareable, crawlable, and independently canonical. Legacy `?article=` links still resolve.
|
|
24
|
+
|
|
7
25
|
## [0.58.0] - 2026-08-17
|
|
8
26
|
|
|
9
27
|
### Added
|
package/README.md
CHANGED
|
@@ -90,7 +90,7 @@ Build the branded one-click bundle:
|
|
|
90
90
|
npm run build:mcpb
|
|
91
91
|
```
|
|
92
92
|
|
|
93
|
-
The generated bundle is written to `build/mcpb/mcp-scraper-<version>.mcpb` and copied to `public/downloads/` for the hosted download. The current public bundle is `https://mcpscraper.dev/downloads/mcp-scraper.mcpb` (`0.
|
|
93
|
+
The generated bundle is written to `build/mcpb/mcp-scraper-<version>.mcpb` and copied to `public/downloads/` for the hosted download. The current public bundle is `https://mcpscraper.dev/downloads/mcp-scraper.mcpb` (`0.60.0`, SHA-256 `a364401f8bc5928735af71bcc254df5b9e64d164e57bfbe103942f5332225f21`). Install it by opening or dragging it into Claude Desktop. Claude displays the `MCP Scraper` install card, icon, and API-key configuration field from the bundle manifest.
|
|
94
94
|
|
|
95
95
|
The MCPB install exposes every tool — web-intelligence plus all `browser_*` tools — through the one `mcp-scraper` server.
|
|
96
96
|
|