Skip to content
verify mcp Beta VerifyMCP is currently in beta. If you notice any issues, get in touch and we’ll put it right.

org.sovgrid/self-hosted-ai

REMOTE · MCP.SOVGRID.ORG · SCANNED SEP 23

Engineering log of self-hosted AI on NVIDIA DGX Spark (GB10/SM121A). 60+ articles indexed.

Available components

0 this week 80 Trust /100
Trust breakdown (7 categories)

How this component scores in each security and reliability category. Every signal is checked automatically against the live server, and we only credit what we can confirm. How we score → Why this is hard to score →

Endpoint Security63
Transport & Reachability100
Schema Quality & AI Usability70
  • AI-judged instruction clarity (excellent).Pass
  • Context-footprint check failed: tool/resource definitions use about 985 tokens (~197/item across 5 items; 5 tools + 0 resources), over budget; trim descriptions and params. See how to fix → Fail
  • Usage-examples check failed: none of the tools include examples. See how to fix → Fail
Stability & Change Management100
  • No destabilizing schema changes in the last 30 days.Pass
Tool Coverage100
  • 100% of tools have a non-trivial description (not blank, and not just the tool's name).Pass
  • 100% of tool parameters carry a description.Pass
  • Structured output schemas are declared (100% of tools); any adoption earns full credit.Pass
Tool Safety100
  • No prompt-injection markers were found in the server instructions, tool names or descriptions we captured.Pass
  • We read all 5 captured tool definition(s), and no name or description among them implies an irreversible operation.Pass
  • An AI judge read all 6 captured unit(s) of tool text and found none that tries to manipulate the model reading it.Pass
Capabilities100
  • Implements a current MCP spec version (2026-07-28).Pass
Install

How do I install the org.sovgrid/self-hosted-ai MCP server?

org.sovgrid/self-hosted-ai is a hosted endpoint at https://mcp.sovgrid.org/self-hosted-ai?ref=mcp-io, so there is nothing to install locally. Ready-made configuration for Claude, Cursor, VS Code, Codex and 5 more is on this page, copied from each client's own documentation.

remote · mcp.sovgrid.org

# add to Claude Code
claude mcp add --transport http org-sovgrid-self-hosted-ai 'https://mcp.sovgrid.org/self-hosted-ai?ref=mcp-io'
// .cursor/mcp.json
{
  "mcpServers": {
    "org-sovgrid-self-hosted-ai": {
      "url": "https://mcp.sovgrid.org/self-hosted-ai?ref=mcp-io"
    }
  }
}
// .vscode/mcp.json
{
  "servers": {
    "org-sovgrid-self-hosted-ai": {
      "type": "http",
      "url": "https://mcp.sovgrid.org/self-hosted-ai?ref=mcp-io"
    }
  }
}
# ~/.codex/config.toml
[mcp_servers.org-sovgrid-self-hosted-ai]
url = "https://mcp.sovgrid.org/self-hosted-ai?ref=mcp-io"
// opencode.json
{
  "$schema": "https://opencode.ai/config.json",
  "mcp": {
    "org-sovgrid-self-hosted-ai": {
      "type": "remote",
      "url": "https://mcp.sovgrid.org/self-hosted-ai?ref=mcp-io",
      "enabled": true
    }
  }
}
# add to OpenClaw
openclaw mcp add org-sovgrid-self-hosted-ai --url 'https://mcp.sovgrid.org/self-hosted-ai?ref=mcp-io' --transport streamable-http
# ~/.hermes/config.yaml
mcp_servers:
  org-sovgrid-self-hosted-ai:
    url: "https://mcp.sovgrid.org/self-hosted-ai?ref=mcp-io"
// ~/.netclaw/config/netclaw.json
{
  "McpServers": {
    "org-sovgrid-self-hosted-ai": {
      "Transport": "http",
      "Url": "https://mcp.sovgrid.org/self-hosted-ai?ref=mcp-io"
    }
  }
}
# add to Vellum
assistant mcp add org-sovgrid-self-hosted-ai -t streamable-http -u 'https://mcp.sovgrid.org/self-hosted-ai?ref=mcp-io'
// mcp.json
{
  "mcpServers": {
    "org-sovgrid-self-hosted-ai": {
      "type": "http",
      "url": "https://mcp.sovgrid.org/self-hosted-ai?ref=mcp-io"
    }
  }
}

The mcpServers block is a cross-client convention. Remote transports vary, so check your client's docs.

Changelog

Every change we have recorded for this component, newest first. Security-relevant changes are always shown. ▲ marks a change for the better, ▼ a change for the worse; unmarked changes are neutral.

  • 22 Sept 26 0
    • MCP protocol: Implements a current MCP spec version (2026-07-28). functional
    • MCP protocol version: 2025-11-25 → 2026-07-28 functional
  • 21 Sept 26 0
    • The server rewrote its instructions, which are the text every model session reads security
    • Schema quality: 823 → 985 functional
    • New tool “list_articles” functional
  • 11 Sept 26 0
    • Injection markers: unverified → pass security
    • Stability: unverified → pass security
    • TLS certificate: unverified → pass security
    • HSTS header: unverified → pass security
    • Transport: fail → pass security
    • Authorization: Authorisation not fully verified: no authorisation is required to call this server, and 4 tool(s) never declared a destructiveHint. The MCP spec treats an absent hint as destructive by default, so we cannot call this surface safe. security
    • Endpoint reachability: unreachable → reachable functional
    • Tool coverage: unverified → 100 functional
    • MCP protocol: unverified → pass functional
  • 10 Sept 26 0
    • Endpoint reachability: reachable → unreachable security
    • Stability: pass → unverified security
    • Tool safety: pass → unverified security
    • TLS certificate: pass → unverified security
    • HSTS header: pass → unverified security
    • Transport: pass → fail security
    • Authorization: Authorisation not yet verified: we couldn't confirm whether this endpoint requires it. security
    • Capabilities: pass → unverified functional
    • Tool coverage: 100 → unverified functional
    • First check of Schema quality: unverified functional
  • 6 Sept 26 +10
    • Transport: fail → pass security
  • 5 Sept 26 −10
    • Transport: pass → fail security
  • 26 Aug 26 +1
    • We updated how we score, so this day's move reflects our rubric, not a change to the server See what changed → functional
  • 25 Aug 26 79
    • Stability: 0.97 → pass security
Diagnostics

Diagnostic detail from the automated scan of this channel: what the scanner observed at each step, so you can see exactly where a check passed or failed. It is informational only and never changes the trust score.

Captured 23 Sept 2026 · Probed https://mcp.sovgrid.org/self-hosted-ai?ref=mcp-io

TLS valid

Negotiated TLS 1.3 with TLS_CHACHA20_POLY1305_SHA256 .

Subject Issuer Valid from Valid until Key Signature Serial
CN=mcp.sovgrid.org CN=YE2,O=Let's Encrypt,C=US 26 Aug 2026 24 Nov 2026 ECDSA 256 ECDSA-SHA384 6de27c22df98c61287a9053d0a9c5011172
SANs: mcp.sovgrid.org
CN=YE2,O=Let's Encrypt,C=US (CA) CN=Root YE,O=ISRG,C=US 3 Sept 2025 2 Sept 2028 ECDSA 384 ECDSA-SHA384 4df3b15dd6c0784c507cd37b58e6f115
CN=Root YE,O=ISRG,C=US (CA) CN=ISRG Root X2,O=Internet Security Research Group,C=US 13 May 2026 2 Sept 2032 ECDSA 384 ECDSA-SHA384 872165fc34b6e5fba8add5b3705fb53a
CN=ISRG Root X2,O=Internet Security Research Group,C=US (CA) CN=ISRG Root X1,O=Internet Security Research Group,C=US 13 May 2026 2 Sept 2032 ECDSA 384 SHA256-RSA 6c8f1dc727c7117f7baf853ac980f9cd

Background: What to check on a remote MCP endpoint →

DNSSEC insecure

Validation of mcp.sovgrid.org. Not signed

Zone DS Keys Algorithms Outcome
. trust_anchor 20326, 38696 8, 8 Verified
org. present 26974 8 Verified
sovgrid.org. absent Unsigned (proven) parent-signed NSEC/NSEC3 proves an unsigned delegation
Authentication No authorisation required

The endpoint answered without asking for a token. Anyone who knows the URL can reach it.

Result No authorisation required
HTTP status 200
Header Value
strict-transport-security max-age=31536000; includeSubDomains
x-content-type-options nosniff
referrer-policy strict-origin-when-cross-origin

Background: How OAuth 2.1 works in the 2026 MCP spec →

Transports 2 probes
Transport URL Outcome Status Location
streamable-http https://mcp.sovgrid.org/self-hosted-ai?ref=mcp-io Verified 200
http (plaintext) http://mcp.sovgrid.org/self-hosted-ai?ref=mcp-io HTTPS enforced 308 https://mcp.sovgrid.org/self-hosted-ai?ref=mcp-io
MCP tools · 5 exposed · ~901 tokens

The tools this component advertises to a client, with an estimated token cost for each. Expand a tool to see its parameters and schema. The per-tool counts are indicative and are not scored directly; the schema's total context footprint is one signal in Schema Quality & AI Usability. A tool's description is untrusted text the model reads on every call, which is what makes this list a security surface and not just an inventory: how tool poisoning works →

Tool Tokens
diagnose_sglang ~299

Validate an SGLang configuration for NVIDIA DGX Spark (GB10/SM121A). Pure pattern-matching against known failure modes documented in the Sovereign AI Blog. No inference, no external calls. Returns critical issues, non-fatal warnings, and a recommended baseline config. All parameters are optional; supply only what you have. With no inputs you get the recommended config and a 'unknown' verdict.

NameTypeReqDescription
attention_backendstringSGLang --attention-backend value (e.g. 'flashinfer', 'triton'). Empty string = skip this check.
cuda_graph_max_bsintegerSGLang --cuda-graph-max-bs value. 0 = skip this check.
error_messagestringPaste error log output here for pattern matching against known failure modes.
hardwarestringHardware description (e.g. 'GB10', 'DGX Spark', 'SM121A'). Empty = skip GB10-specific rules.
image_tagstringDocker image tag in use (e.g. 'lmsysorg/sglang:latest', 'lmsysorg/sglang:v0.4.0'). Empty = skip.
mem_fractionnumberSGLang --mem-fraction-static value (e.g. 0.88). 0.0 = skip this check.
NameTypeReqDescription
issuesarrayyesCritical issues that will prevent SGLang from running correctly
recommended_configyesVerified-good baseline config for GB10/SM121A
verdictstringyesOverall verdict. 'unknown' = no inputs provided.
warningsarrayyesNon-fatal warnings (suboptimal but non-blocking)

No examples provided.

get_article ~99

Retrieve the full content of a blog article by its slug. Returns the article body (Markdown) plus metadata. If the slug does not match any article, returns an Article with `error='article_not_found'` and other fields at their defaults.

NameTypeReqDescription
slugstringyesArticle slug as returned by search_blog (e.g. 'setup-llm-inference-setup'). Lower-case, hyphenated.
NameTypeReqDescription
bodystringFull article body in Markdown
datestringPublication date (ISO 8601)
descriptionstringShort article description
errorSet to 'article_not_found' if no article matches the slug
quality_classstringEditorial content class (e.g. 'Ephemeral', 'Evergreen'). Empty if not classified.
quality_scorenumberBuild-time quality score from the editorial pipeline (unbounded weighted composite across 13 signals, higher is better; thresholds depend on style)
quality_stylestringEditorial style category (e.g. 'best_practice_learnings', 'werthaltige_code_beispiele'). Empty if not categorised.
slugstringyesArticle slug
tagsarrayTopic tags assigned to the article
titlestringArticle title
urlstringPublic URL of the article
word_countintegerWord count of the article body

No examples provided.

list_articles ~158

List all blog articles. No TF-IDF computation — pure database listing. Use to browse the full corpus, paginate through articles, or filter by tag. For full-text semantic search use search_blog instead.

NameTypeReqDescription
limitintegerNumber of results (1-50)
offsetintegerPagination offset (0-based)
sortstringResult ordering. 'date_desc' newest first (default). 'date_asc' oldest first. 'title_asc' alphabetical. 'quality_desc' best quality first.
tagOptional tag filter (e.g. 'setup', 'fixes', 'strategy'). Only articles with this tag are considered. Use list_tags to discover available tags.
NameTypeReqDescription
resultarrayyes

No examples provided.

list_tags ~76

List all topic tags used across the Sovereign AI Blog corpus, with article counts. Use this to browse the topic space before calling search_blog with a tag filter.

NameTypeReqDescription
sortstringResult ordering. 'count_desc' lists most-used tags first (default). 'alpha' sorts alphabetically.
NameTypeReqDescription
resultarrayyes

No examples provided.

search_blog ~269

Search the Sovereign AI Blog for articles matching a natural language query, optionally filtered by tag and sorted by relevance or date. Behaviour matrix: - query='', sort=* -> list newest-first, optionally tag-filtered - query!='', sort=relevance -> TF-IDF ranked, optionally tag-filtered - query!='', sort=date_desc -> TF-IDF filtered (score > 0.001), then sorted by date Pure read-only, deterministic for a given KB snapshot.

NameTypeReqDescription
nintegerMaximum number of results to return
querystringNatural language search query (e.g. 'flashinfer OOM on GB10'). Multi-word queries are tokenized and TF-IDF ranked. Pass empty string to list articles without ranking by relevance.
sortstringResult ordering. 'relevance' uses TF-IDF score (default for non-empty query). 'date_desc' sorts newest first (default behaviour when query is empty). When query is empty, 'relevance' is treated as 'd…
tagOptional tag filter (e.g. 'setup', 'fixes', 'strategy'). Only articles with this tag are considered. Use list_tags to discover available tags.
NameTypeReqDescription
resultarrayyes

No examples provided.

Common questions

What is the org.sovgrid/self-hosted-ai MCP server?

org.sovgrid/self-hosted-ai is an MCP server listed in the public MCP registry as org.sovgrid/self-hosted-ai. Engineering log of self-hosted AI on NVIDIA DGX Spark (GB10/SM121A). 60+ articles indexed. This page covers its hosted endpoint (https://mcp.sovgrid.org/self-hosted-ai?ref=mcp-io).

Is the org.sovgrid/self-hosted-ai MCP server safe to use?

org.sovgrid/self-hosted-ai scores 80 out of 100 on VerifyMCP. That is a record of what we were able to check automatically, not an endorsement. The category breakdown on this page shows every signal behind the number, including the ones we could not confirm.

What tools does the org.sovgrid/self-hosted-ai MCP server expose?

org.sovgrid/self-hosted-ai exposes 5 tools: search_blog, list_articles, get_article, diagnose_sglang, list_tags. Their descriptions and schemas cost roughly 901 tokens of context every time the server is loaded.

Does the org.sovgrid/self-hosted-ai MCP server require authentication?

No. We connected to org.sovgrid/self-hosted-ai without credentials and it answered, so anything it exposes is reachable by anyone who knows the address.

Is the org.sovgrid/self-hosted-ai MCP server still maintained?

org.sovgrid/self-hosted-ai is still listed as active in the MCP registry. We last reached this channel on 23 September 2026. Those dates come from our own scans of the registry and the channel itself, not from anything the publisher announced.