org.sovgrid/self-hosted-ai

Measured conclusion: Deep-tested. L2 outcome: tools_ok.

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

Add to your agent

Copy this into your MCP client config:

{
  "mcpServers": {
    "self-hosted-ai": {
      "url": "https://mcp.sovgrid.org/self-hosted-ai?ref=mcp-io",
      "transport": "streamable-http"
    }
  }
}

Measured tools (5)

• search_blog read-only

  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.

  Parameters:

  • n — integer — Maximum number of results to return
  • tag — Optional tag filter (e.g. 'setup', 'fixes', 'strategy'). Only articles with this tag are considered. Use list_tags to discover available tags.
  • sort — string — Result 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 'date_desc'.
  • query — string — Natural 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.

  Hints: read-only yes · destructive not declared · idempotent yes · open-world no

• list_articles read-only

  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.

  Parameters:

  • tag — Optional tag filter (e.g. 'setup', 'fixes', 'strategy'). Only articles with this tag are considered. Use list_tags to discover available tags.
  • sort — string — Result ordering. 'date_desc' newest first (default). 'date_asc' oldest first. 'title_asc' alphabetical. 'quality_desc' best quality first.
  • limit — integer — Number of results (1-50)
  • offset — integer — Pagination offset (0-based)

  Hints: read-only yes · destructive not declared · idempotent yes · open-world no

• get_article read-only

  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.

  Parameters:

  • slug (required) — string — Article slug as returned by search_blog (e.g. 'setup-llm-inference-setup'). Lower-case, hyphenated.

  Hints: read-only yes · destructive not declared · idempotent yes · open-world no

• diagnose_sglang read-only

  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.

  Parameters:

  • hardware — string — Hardware description (e.g. 'GB10', 'DGX Spark', 'SM121A'). Empty = skip GB10-specific rules.
  • image_tag — string — Docker image tag in use (e.g. 'lmsysorg/sglang:latest', 'lmsysorg/sglang:v0.4.0'). Empty = skip.
  • mem_fraction — number — SGLang --mem-fraction-static value (e.g. 0.88). 0.0 = skip this check.
  • error_message — string — Paste error log output here for pattern matching against known failure modes.
  • attention_backend — string — SGLang --attention-backend value (e.g. 'flashinfer', 'triton'). Empty string = skip this check.
  • cuda_graph_max_bs — integer — SGLang --cuda-graph-max-bs value. 0 = skip this check.

  Hints: read-only yes · destructive not declared · idempotent yes · open-world no

• list_tags read-only

  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.

  Parameters:

  • sort — string — Result ordering. 'count_desc' lists most-used tags first (default). 'alpha' sorts alphabetically.

  Hints: read-only yes · destructive not declared · idempotent yes · open-world no

Publisher website Repository

Measured by MCPYP probing. Untested fields are omitted, never estimated. Listing is always free; MCPYP never proxies calls or stores user tokens.