MCP
需 API Key
project-summarizer
x402 Text Summarizer compresses text, Markdown, or JSON into structured summaries using Claude AI
原作者:gary-edgington
来源:Smithery
仅卡片存档
目录 / 文档处理 / Meshnote
A persistent Markdown wiki that your AI agent maintains as long-term memory across sessions.
https://meshnote--b-nicovski.run.tools{
"mcpServers": {
"Meshnote": {
"headers": {
"Authorization": "Bearer \u003cYOUR_KEY\u003e"
},
"url": "https://meshnote--b-nicovski.run.tools"
}
}
}
| 工具 | 说明 |
|---|---|
| append_log | Append an entry to wiki/log.md with a parseable header. |
| get_index | Return the contents of wiki/index.md — the LLM-maintained catalog of the wiki. |
| list_page_history | List saved versions of a wiki page. Previous content is auto-saved before every overwrite. |
| get_backlinks | Return every page that links to the given target path. |
| read_page | Return a wiki page's full content, frontmatter, metadata, and version token. Hand the returned 'version' back to write_page as 'base_version' so concurrent writers can't silently clobber each other. |
| list_pages | Enumerate wiki pages, optionally filtered by kind (entity, concept, summary, synthesis, index, log). |
| search_wiki | BM25 search across wiki pages. Returns the top hits ordered by relevance. |
| list_sources | Enumerate raw sources under raw/sources/. |
| get_graph | Return the full wiki link graph as nodes and edges, for synthesis operations. |
| get_schema | Return the contents of the project's schema.md. An agent should call this first to learn how the user wants the wiki maintained. |
| create_project | Create a new meshnote project (brain) with the given name and template. Templates: custom, research, book, personal, business. |
| read_source | Return the full content of a raw source file. |
| list_projects | List all meshnote projects registered on this machine. |
| restore_page | Restore a wiki page to a previous version. The current content is saved to history first, so nothing is lost. Use list_page_history to find available versions. Bypasses the base_version check by design — restore is an explicit overwrite. |
| write_page | Create or overwrite a wiki page atomically. Path must live under wiki/. Use base_version to prevent silently clobbering a concurrent edit: - To UPDATE an existing page: call read_page first and pass its 'version' as base_version. If another writer modified the page in between, the call fails with a CONFLICT error; re-read and retry. - To CREATE a new page: pass base_version="" (empty string) to assert the page does not exist yet. (If base_version is omitted entirely, a new page is still created; only updating an existing page requires the token.) Previous content is auto-saved to _history/ — use list_page_history to see saved versions. The result lists any 'unresolved_links' — wikilinks on this page that match no existing page by title, path or slug. That is a warning, not an error: linking to a page you have not written yet is fine. Check it for typos before moving on. |
| register_source | Download a URL and register it as a raw source. Does not call any LLM — the agent decides what to do next. When preserveOriginal is true, the bytes are saved verbatim with the original extension (e.g. paper.pdf) plus a sidecar paper.pdf.meta.md with frontmatter. |
x402 Text Summarizer compresses text, Markdown, or JSON into structured summaries using Claude AI
A markdown workspace that becomes a shared brain for you and your AI agents — 25 tools to read, write, organize, and sha…
Convert PDF, Word, PowerPoint, Excel, HTML and EPUB files into clean Markdown for LLMs, including OCR for scanned PDFs a…
PDF Toolkit converts PDFs/images to Word, Excel, and Markdown, splits and merges PDFs, creates PDFs from images, adds te…
# MinerU Document to Markdown
2
侵权举报经核验成立后,我们会即时下线该条目并删除已存的内容副本。