Skip to content

corpus MCP server

study.vela/corpusvela.studystreamable-httpv0.1.0Scanned

Read-only MCP access to Vela's corpus of human experience: cited passages, coordinates, reading path

10 tools: 4 change things and 6 only read.

Tools: 0 delete, 0 irreversible, 4 write, 6 read.
  • Reads private data · 0
  • Reads text others write · 0
  • Can send data out · 0
  • No findings

Tools listed without sign-in. Remote posture

Anlyon alternative

4 of these tools change something. Run those as governed actions.

Declare what an action changes, and Anlyon puts it under a shared limit, binds every approval to the exact request, and grades the outcome. It works on any HTTPS API you can describe as an action.

Anlyon Radar does not run, proxy or govern this MCP server. A tool your own code calls directly is not routed through Anlyon and is not governed by it.

Import an OpenAPI spec →

What the corpus MCP server can do

The corpus MCP server is published by vela.study, a domain the MCP Registry verified. It has 10 tools. It can create or change data (4 tools: vela.corpus.ground_claim, vela.corpus.witness, vela.magazine.sequence_path and vela.magazine.relation_floor), but cannot delete data. 6 tools only read.

None of its tools reads private data, brings in text other people write, or sends data out.

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.

Last scanned Nov 4, 2026, version 0.1.0.

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.

10 tools

Changes things 4Claude Code: ask
  • vela.corpus.ground_claimGround a claim in the corpus: hybrid semantic + theme retrieval over real passages, returning citations with source, passage code, scores, and a SHORT snippet.No findings

    Why it is in this group

    • Nothing in its name or description says what it does, so it is counted as a write.
    • The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
    Description, as the server wrote it
    Ground a claim in the corpus: hybrid semantic + theme retrieval over real passages, returning citations with source, passage code, scores, and a SHORT snippet. Use to check whether the corpus supports, complicates or is silent on an assertion. COSTS a small amount (one embedding per call). [capability: corpus-ingestion-dual-grade · mirrors: internal — lib/audit/retrieve.ts]

    Annotations the server declares

    • readOnlyHint: true
    • destructiveHint: false
    • openWorldHint: false
  • vela.corpus.witnessLiterary Witness Rail (ASN-1577): return passages and one-line claims from Vela's 762-work corpus that illuminate the given themes and/or emotions.No findings

    Why it is in this group

    • Nothing in its name or description says what it does, so it is counted as a write.
    • The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
    Description, as the server wrote it
    Literary Witness Rail (ASN-1577): return passages and one-line claims from Vela's 762-work corpus that illuminate the given themes and/or emotions. Designed for frame-family profile pages — PersonFrame × ProblemFrame × ConditionFrame — to add an inside-out layer (what writers who lived this say) alongside the outside-in portrait the frame join produces. Results carry evidence_grade 'T1' (single-extractor) until PRN-282 lands. FREE for claims; costs one embedding for the passage path. [capability: literary-witness · mirrors: GET /api/vela/witness]

    Annotations the server declares

    • readOnlyHint: true
    • destructiveHint: false
    • openWorldHint: false
  • vela.magazine.relation_floorWhether a magazine's corpus can support the adaptive promise its pages render.No findings

    Why it is in this group

    • Nothing in its name or description says what it does, so it is counted as a write.
    • The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
    Description, as the server wrote it
    Whether a magazine's corpus can support the adaptive promise its pages render. Reports, per property, each relational dimension's fill AND its separating power — a field present on every piece with one distinct value carries no information, so fill rate alone is misleading. Read-only. [capability: adaptive-magazine-engine · mirrors: npm run magazine:relation-floor]

    Annotations the server declares

    • readOnlyHint: true
    • destructiveHint: false
    • openWorldHint: false
  • vela.magazine.sequence_pathAsk what this magazine would actually give a reader who said they wanted these things.No findings

    Why it is in this group

    • Nothing in its name or description says what it does, so it is counted as a write.
    • The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
    Description, as the server wrote it
    Ask what this magazine would actually give a reader who said they wanted these things. Returns the REAL sequenced path with the rationale for each piece — the same core the anonymous intake calls, not a preview. Read-only: writes nothing and records no signal. [capability: adaptive-magazine-engine · mirrors: POST /api/magazine/start-path]

    Annotations the server declares

    • readOnlyHint: true
    • destructiveHint: false
    • openWorldHint: false
Only reads 6Show 6 read-only tools
  • vela.corpus.describe_topicDescribe ONE topic in full — its axes and their operative definitions, its complete controlled theme and quality vocabularies, plus the vocabulary actually IN USE across the corpus…No findings

    Why it is in this group

    • The name "vela.corpus.describe_topic" says "describe": a word for reading.
    Description, as the server wrote it
    Describe ONE topic in full — its axes and their operative definitions, its complete controlled theme and quality vocabularies, plus the vocabulary actually IN USE across the corpus with counts. The declared vocabulary and the used vocabulary are different facts; this returns both. FREE and read-only. [capability: topic-coordinate-systems · mirrors: GET /api/corpus/topics/:slug]

    Annotations the server declares

    • readOnlyHint: true
    • destructiveHint: false
    • openWorldHint: false
  • vela.corpus.list_topicsList the coordinate systems the corpus can be queried through — each with its thesis, axes, and controlled vocabularies.No findings

    Why it is in this group

    • The name "vela.corpus.list_topics" says "list": a word for reading.
    Description, as the server wrote it
    List the coordinate systems the corpus can be queried through — each with its thesis, axes, and controlled vocabularies. Call this before vela.corpus.query to learn what axes and themes exist. FREE and read-only. [capability: topic-coordinate-systems · mirrors: GET /api/corpus/topics]

    Annotations the server declares

    • readOnlyHint: true
    • destructiveHint: false
    • openWorldHint: false
  • vela.corpus.queryAsk the corpus by COORDINATE rather than keyword.No findings

    Why it is in this group

    • The name "vela.corpus.query" says "query": a word for reading.
    Description, as the server wrote it
    Ask the corpus by COORDINATE rather than keyword. Filter works by axis weights (e.g. religion ∈ {T,s} AND work ∈ {T,s} returns the God-and-Work source set across genres), controlled themes, qualities/emotions, genre, and free text. Returns card metadata: the one-line claim, themes, coordinates, qualities. FREE and read-only. [capability: library-extraction-cards · mirrors: POST /api/corpus/query]

    Annotations the server declares

    • readOnlyHint: true
    • destructiveHint: false
    • openWorldHint: false
  • vela.library.reading_listThe reading list an emotion or theme page shows: books with one-line claims and a short attributed passage (≤300 chars).No findings

    Why it is in this group

    • The name "vela.library.reading_list" says "list": a word for reading.
    Description, as the server wrote it
    The reading list an emotion or theme page shows: books with one-line claims and a short attributed passage (≤300 chars). Exactly one of emotion or theme. FREE and read-only. Every url carries ?ref=mcp for citation arrival measurement. [capability: library-extraction-cards · mirrors: GET /api/corpus/reading-list]

    Annotations the server declares

    • readOnlyHint: true
    • destructiveHint: false
    • openWorldHint: false
  • vela.research.get_studyOne FieldStation study as a library-kit ProfileDTO plus the raw StudyRecord.No findings

    Why it is in this group

    • The name "vela.research.get_study" says "get": a word for reading.
    Description, as the server wrote it
    One FieldStation study as a library-kit ProfileDTO plus the raw StudyRecord. Unknown id is an error. FREE and read-only. [capability: study-catalogue · mirrors: GET /api/research/studies/:id]

    Annotations the server declares

    • readOnlyHint: true
    • destructiveHint: false
    • openWorldHint: false
  • vela.research.list_studiesList FieldStation research ideas as a library-kit DiscoveryDTO.No findings

    Why it is in this group

    • The name "vela.research.list_studies" says "list": a word for reading.
    Description, as the server wrote it
    List FieldStation research ideas as a library-kit DiscoveryDTO. Each card is a study we can support, not a running participant surface. The honesty gate (limitations, readiness, doNotClaim) is enforced by research:catalog-check — this tool publishes that same file. FREE and read-only. Presence is gated; truth is not. [capability: study-catalogue · mirrors: GET /api/research/studies]

    Annotations the server declares

    • readOnlyHint: true
    • destructiveHint: false
    • 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 private-data or untrusted-input or outbound.

Reads private data0
Reads files, mail, messages, databases, secrets, calendars or repositories.
Reads text others write0
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.

Tools listed without sign-in

Looked at Nov 20, 2026, 19:19 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
4
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
  • WarnThe whole server

    The endpoint served tools/list without credentials, and the list has 4 tools that change things (4 write). Whether a call also works without credentials was not tested: tools are never called.

    Matched text

    https://vela.study/mcp: tools/list returned 10 tools to an anonymous client, including vela.corpus.ground_claim (write), vela.corpus.witness (write), vela.magazine.sequence_path (write), vela.magazine.relation_floor (write)

    What this check looks for

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.

No change seen yet. Its tools are rescanned every day, and a remote endpoint is looked at once a week.

Install

Pinned to version 0.1.0. Values in angle brackets are yours to fill in.

claude mcp add --transport http corpus https://vela.study/mcp
claude mcp add --transport http corpus https://vela.study/mcp

Run 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.

.claude/settings.json
{
  "permissions": {
    "allow": [
      "mcp__corpus__vela_corpus_query",
      "mcp__corpus__vela_library_reading_list",
      "mcp__corpus__vela_corpus_list_topics",
      "mcp__corpus__vela_corpus_describe_topic",
      "mcp__corpus__vela_research_list_studies",
      "mcp__corpus__vela_research_get_study"
    ],
    "ask": [
      "mcp__corpus__vela_corpus_ground_claim",
      "mcp__corpus__vela_corpus_witness",
      "mcp__corpus__vela_magazine_sequence_path",
      "mcp__corpus__vela_magazine_relation_floor"
    ],
    "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 corpus. Review it before you use it.

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/study.vela/corpus/verify?hash=629b6db7d31e8d0823bc40f9ec78e1b9cd99f1f9255edce1c5f1e8cbbb03d60a" | jq -e '.changed == false'

Pinned hash 629b6db7d31e, from Nov 4, 2026, 15:18 UTC.

Watch

Get told when its tools change: a new tool, a new description, a new class.

https://radar.anlyon.com/servers/study.vela/corpus/changes.xml

An Atom feed. Works in any feed reader or Slack's RSS app.

Badge

Anlyon Radar badge for study.vela/corpus
[![Anlyon Radar](https://radar.anlyon.com/badge/study.vela/corpus.svg)](https://radar.anlyon.com/servers/study.vela/corpus)

Source: MCP Registry · repository. Synced Oct 21, 2026, 18:55 UTC.

corpus MCP server: questions and answers

Answered from the scan above, not from the publisher's description.

Is the corpus MCP server safe?

No scan can promise that. What its definitions show: 10 tools, 4 of which can change, send or delete something, and 0 findings from fixed checks. Allow the reads, ask before each change, and deny the deletes; the permissions block on this page does that.

Can the corpus MCP server delete data?

No tool is classed as destructive: going by their definitions, none of them deletes, cancels or revokes anything.

Can the corpus 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 corpus MCP server have?

10 tools: 4 write, 6 read. vela.corpus.ground_claim, vela.corpus.witness, vela.magazine.sequence_path, vela.magazine.relation_floor, vela.corpus.query, vela.library.reading_list, vela.corpus.list_topics, vela.corpus.describe_topic and 2 more.

Does the corpus MCP server need sign-in or OAuth?

Its remote endpoint listed its tools to an anonymous client, without sign-in.

Does the corpus MCP server send data to outside services?

None of its tools sends, posts or publishes, or takes a recipient, channel or URL.

Has the corpus MCP server changed its tools?

Not since Anlyon Radar started tracking it. Last scanned Nov 4, 2026.