Vedaword MCP server
com.vedaword/libraryvedaword.comstreamable-httpv1.1.0Scanned
Sanskrit scripture search: Mahabharata, Upanishads, Gita; cited verses, honest no-answer.
8 tools: 8 only read.
- Reads private data · 2
- Reads text others write · 1
- Can send data out · 0
- No findings
Tools listed without sign-in. Remote posture
With Anlyon
Decide which of this server's tools your agent can use.
- Anlyon lists the tools and adopts none.
- Each tool you adopt gets approvals and receipts.
- Your agent never holds the key.
A tool your agent calls directly is not routed through Anlyon and is not governed by it. Anlyon Radar only lists and labels, and these labels are the same either way.
What the Vedaword MCP server can do
The Vedaword MCP server is published by vedaword.com, a domain the MCP Registry verified. It has 8 tools. Every one of them only reads: going by their definitions, none of them creates, changes, sends or deletes anything.
2 tools read private data (get_vaishnava_calendar and list_authors); 1 tool brings in text other people can write (browse_work).
None of the fixed checks matched anything in its tool definitions. That is not a safety guarantee.
Its remote endpoint lists its tools without sign-in, and its tools changed once since Anlyon Radar first saw them.
Last scanned Nov 12, 2026, version 1.1.0.
Tools
Grouped by what each tool does, most consequential first. Open one for its reasons and full description; descriptions are the publisher's own words.
8 tools
Only reads 8Show 8 read-only tools
browse_workWalk a work's structure in order: a work gives its sections, a section its chapters, a chapter its verses (label and translation, in pages).Reads text others writeNo findings
Why it is in this group
- The name "browse_work" says "browse": a word for reading.
What it can reach
- untrusted-inputThe name "browse_work" says "browse": it reads web pages, which anyone can write.
Description, as the server wrote it Walk a work's structure in order: a work gives its sections, a section its chapters, a chapter its verses (label and translation, in pages). Use it when the user wants to read a chapter or a book through, asks what comes after a verse, or what a book contains; search_texts cannot do that. Three steps, each named in the reply of the one before: `work` alone lists the sections; add `section` for its chapters; add `chapter` (the `index` from the listing) for its verses. Every reply carries `next`, the arguments of the call that continues. To quote a verse, call get_verse with its `unit_id`: it returns the citation and version hash.
Annotations the server declares
- readOnlyHint: true
- openWorldHint: false
find_similar_versesGiven a verse, find other passages in the library that say something close to it — by meaning, not by shared words.No findings
Why it is in this group
- The name "find_similar_verses" says "find": a word for reading.
Description, as the server wrote it Given a verse, find other passages in the library that say something close to it — by meaning, not by shared words. Start here when you already have a verse; when you only have words, a question or a term, use search_texts. Use it to check whether a teaching appears elsewhere before presenting it as the tradition's position, to gather parallel places, or to see how different works treat one subject. Set `exclude_same_work` when you want an independent witness: the next verse of the same chapter is nearly always similar and proves nothing. An empty result is a real answer — it means the library says this in one place only.
Annotations the server declares
- readOnlyHint: true
- openWorldHint: false
get_verseFetch one verse whole: the Sanskrit in Devanāgarī and IAST, the word-by-word breakdown with the meaning each word carries in this passage, the literary translation, the commentary,…No findings
Why it is in this group
- The name "get_verse" says "get": a word for reading.
Description, as the server wrote it Fetch one verse whole: the Sanskrit in Devanāgarī and IAST, the word-by-word breakdown with the meaning each word carries in this passage, the literary translation, the commentary, and optionally the surrounding verses. Call it after search_texts when you are going to quote, and whenever the user names a verse. The word-by-word table is what lets you explain or re-translate a line honestly instead of trusting a finished translation. Accepts citations as people write them, not only internal ids. To read a chapter or a whole book in order, use browse_work. To read on around one verse, call this again with `context` 5 and then with the `unit_id` of the last neighbour marked "after"; each reply carries the five verses on either side with their translations.
Annotations the server declares
- readOnlyHint: true
- openWorldHint: false
list_authorsList the authors the tradition names, grouped by period, each with a short note and the books credited to them.Reads private dataNo findings
Why it is in this group
- The name "list_authors" says "list": a word for reading.
What it can reach
- private-dataThe description's first sentence says "list" and "note": it reads notes.
Description, as the server wrote it List the authors the tradition names, grouped by period, each with a short note and the books credited to them. Call it when the user asks who wrote something, or wants the works of one acharya; pass the author's `id` to list_works to see the books. Filter by `era`: shruti, classical, caitanya, successors, modern.
Annotations the server declares
- readOnlyHint: true
- openWorldHint: false
list_worksList the works with their sizes, shelf, author and available languages.No findings
Why it is in this group
- The name "list_works" says "list": a word for reading.
Description, as the server wrote it List the works with their sizes, shelf, author and available languages. Call it at the start of a conversation about scripture, or when you need work slugs for filtering. The list is short by default: no descriptions. Narrow it with `category` (a shelf: itihasa, purana, upanishad, vedanta, samhita, shastra, kavya, shruti, smriti; `shruti` is the upanishad shelf, `smriti` the itihasa and purana shelves) or with `author` (an id from list_authors). Set `include_descriptions` for a paragraph on each work. It also tells you what is not translated: several works exist only in Russian. Check `languages` before promising the user an English text.
Annotations the server declares
- readOnlyHint: true
- openWorldHint: false
lookup_sanskrit_termLook up a Sanskrit word: its spellings across scripts, the meanings it carries in this corpus, how many verses gloss it, and examples of where it occurs.No findings
Why it is in this group
- The name "lookup_sanskrit_term" says "lookup": a word for reading.
Description, as the server wrote it Look up a Sanskrit word: its spellings across scripts, the meanings it carries in this corpus, how many verses gloss it, and examples of where it occurs. Built from the word-by-word breakdowns of some 30 000 verses, so these are the meanings translators actually assigned to the word in these texts — not a lexicon's list of everything it could mean. Use it before translating a term yourself, and whenever the user asks what a word means.
Annotations the server declares
- readOnlyHint: true
- openWorldHint: false
search_textsSearch the Vedaword library of Sanskrit scripture by meaning as well as by wording, so a paraphrased question works and a Sanskrit term reaches passages written in another script.No findings
Why it is in this group
- The name "search_texts" says "search": a word for reading.
Description, as the server wrote it Search the Vedaword library of Sanskrit scripture by meaning as well as by wording, so a paraphrased question works and a Sanskrit term reaches passages written in another script. This is the first thing to call for any factual question about what the scriptures say. Start here when you have words: a question, a phrase, a term. When you already have a verse and want its parallels, use find_similar_verses instead; to read a known verse whole, use get_verse; to read a chapter or a book in order, use browse_work. Returns passages with an exact citation, a URL and the version hash of the text quoted. Read `confidence` before using the results. "none" means the library has no answer and the result list is empty — report that to the user instead of reaching for something related. "low" means the passages are leads, not answers. `matched_by` says how each was found: a passage found only by "vector" shares no words with the query.
Annotations the server declares
- readOnlyHint: true
- openWorldHint: false
This is not a safety rating, and no finding is not a safety guarantee. The groups and findings come from fixed checks of the tool definitions the server returned on the date shown, not from its code or what it does when called. How the checks work.
Exfil surface
Which tools read private data, which read text someone else can write, and which can send data out. One of each is a path a prompt injection can use.
No private-data, untrusted-input, outbound path in this server alone: no tool is tagged outbound.
- Reads private data2
- Reads files, mail, messages, databases, secrets, calendars or repositories.
- Reads text others write1
- Brings in text someone else can write: web pages, inbound email, messages, issues and comments.
- Can send data out0
- Can carry data to people or systems: it sends, posts or publishes, or takes a recipient, channel or URL.
Other servers in the same agent add their own tools to these paths. Audit your whole setup to see them together.
Remote posture
What anyone on the internet sees at the remote endpoint: an anonymous initialize and tools/list, and the public OAuth metadata.
Looked at Nov 20, 2026, 18:49 UTC, unchanged since Oct 22, 2026. No credentials were sent and no tool was called. What is looked at.
- Reachable
- yes
- HTTP status of an anonymous initialize
- 200
- tools/list served without credentials
- yes
- Tools that change things, listed without credentials
- 0
- Protocol version
- 2025-06-18
- WWW-Authenticate on 401
- not observed
- resource_metadata in WWW-Authenticate
- not observed
- RFC 9728 protected resource metadata
- not observed
- Authorization servers
- not observed
- Legacy HTTP+SSE transport
- no
- Dynamic client registration
- not observed
- Client ID metadata documents
- not observed
Findings
Matches of fixed checks, with the text they matched.
No check matched anything in these tool definitions. That is not a safety guarantee.
History
Changes to its tool definitions and to its remote endpoint, newest first.
Tools8 changed.~8 changedView diff
- Changed:browse_workOnly readsinputSchema
- Changed:find_similar_versesOnly readsinputSchema
- Changed:get_vaishnava_calendarOnly readsinputSchema
- Changed:get_verseOnly readsinputSchema
- Changed:list_authorsOnly readsinputSchema
- Changed:list_worksOnly readsinputSchema
- Changed:lookup_sanskrit_termOnly readsinputSchema
- Changed:search_textsOnly readsinputSchema
Install
Pinned to version 1.1.0. Values in angle brackets are yours to fill in.
claude mcp add --transport http library https://mcp.vedaword.comclaude mcp add --transport http library https://mcp.vedaword.comRun in a terminal.
Claude Code permissions
Allow the reads, ask before a change or a way out, deny the deletes. Paste into .claude/settings.json.
{
"permissions": {
"allow": [
"mcp__library__search_texts",
"mcp__library__get_verse",
"mcp__library__lookup_sanskrit_term",
"mcp__library__find_similar_verses",
"mcp__library__get_vaishnava_calendar",
"mcp__library__list_works",
"mcp__library__list_authors",
"mcp__library__browse_work"
],
"ask": [],
"deny": []
}
}Tools that only read are allowed, tools that change things ask first, tools that delete or cancel are denied. A read that can send data out (it takes a URL, say) asks first too. It assumes the server is installed as library. Review it before you use it.
Capability tags behind 3 rules
- mcp__library__get_vaishnava_calendar private-data
- mcp__library__list_authors private-data
- mcp__library__browse_work untrusted-input
Pin
Keep the decision: fail CI when the tools change, follow every change, or show a badge.
Pin in CI
Fail a build when the tools you reviewed are not the tools the server serves today.
curl -s "https://radar.anlyon.com/api/v1/servers/com.vedaword/library/verify?hash=48ee051563e5475c73de9daca33ef88e980d3b5d41f4c39ba96448e25ec3801a" | jq -e '.changed == false'Pinned hash 48ee051563e5, from Nov 12, 2026, 21:17 UTC.
Watch
Get told when its tools change: a new tool, a new description, a new class.
https://radar.anlyon.com/servers/com.vedaword/library/changes.xmlAn Atom feed. Works in any feed reader or Slack's RSS app.
Badge
[](https://radar.anlyon.com/servers/com.vedaword/library)Source: MCP Registry. Synced Oct 31, 2026, 02:17 UTC.
Vedaword MCP server: questions and answers
Answered from the scan above, not from the publisher's description.
Is the Vedaword MCP server safe?
- No scan can promise that. What its definitions show: 8 tools, all of which only read, and 0 findings from fixed checks. Read-only tools can still pass private data to a model.
Can the Vedaword MCP server delete data?
- No tool is classed as destructive: going by their definitions, none of them deletes, cancels or revokes anything.
Can the Vedaword MCP server do things that cannot be undone?
- No tool is classed as irreversible: none of them sends, pays, publishes or merges, going by their definitions.
What tools does the Vedaword MCP server have?
- 8 tools: 8 read. search_texts, get_verse, lookup_sanskrit_term, find_similar_verses, get_vaishnava_calendar, list_works, list_authors and browse_work.
Does the Vedaword MCP server need sign-in or OAuth?
- Its remote endpoint listed its tools to an anonymous client, without sign-in.
Does the Vedaword MCP server send data to outside services?
- None of its tools sends, posts or publishes, or takes a recipient, channel or URL.
Has the Vedaword MCP server changed its tools?
- Yes, once since Anlyon Radar started tracking it, most recently on Oct 31, 2026. Every change is on this page and in its Atom feed.