org.sovgrid/self-hosted-ai

Est-ce que org.sovgrid/self-hosted-ai est sûr ?

Au dernier audit MCP Hub, le serveur MCP « org.sovgrid/self-hosted-ai » est classé « Non encore testé ». Référencé dans l'annuaire. Seul le pentest comportemental fait foi, et il n'a pas encore été exécuté sur ce serveur. Le verdict est réévalué en continu ; consultez la fiche pour l'état à jour.

Méthode : MCP Hub analyse la description de chaque outil (couche 1) et, pour le verdict fort, exécute le serveur dans une micro-VM isolée avec leurres et réseau bloqué (couche 2). On publie des noms et des compteurs, jamais le contenu brut renvoyé par un outil.

Les 5 outils

• diagnose_sglang≈ 1180 tokens

  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.

  Paramètres, retour & exemple (6)

  Paramètres

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

  Exemple d'appel

  {"name": "diagnose_sglang", "arguments": {"hardware": "texte", "image_tag": "texte", "mem_fraction": 1, "error_message": "texte", "attention_backend": "texte", "cuda_graph_max_bs": 1}}

  Exemple de retour

  {"content": [{"type": "text", "text": "…"}], "structuredContent": {"issues": [], "verdict": "valid", "warnings": [], "recommended_config": "texte"}}

• get_article≈ 623 tokens

  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.

  Paramètres, retour & exemple (1)

  Paramètres

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

  Exemple d'appel

  {"name": "get_article", "arguments": {"slug": "texte"}}

  Exemple de retour

  {"content": [{"type": "text", "text": "…"}], "structuredContent": {"url": "texte", "body": "texte", "date": "texte", "slug": "texte", "tags": [], "error": "texte", "title": "texte", "word_count": 1, "description": "texte", "quality_class": "texte", "quality_score": 1, "quality_style": "texte"}}

• list_articles≈ 666 tokens

  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.

  Paramètres, retour & exemple (4)

  Paramètres

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

  Exemple d'appel

  {"name": "list_articles", "arguments": {"tag": "texte", "sort": "date_desc", "limit": 1, "offset": 1}}

  Exemple de retour

  {"content": [{"type": "text", "text": "…"}], "structuredContent": {"result": []}}

• list_tags≈ 308 tokens

  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.

  Paramètres, retour & exemple (1)

  Paramètres

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

  Exemple d'appel

  {"name": "list_tags", "arguments": {"sort": "count_desc"}}

  Exemple de retour

  {"content": [{"type": "text", "text": "…"}], "structuredContent": {"result": []}}

• search_blog≈ 796 tokens

  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.

  Paramètres, retour & exemple (4)

  Paramètres

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

  Exemple d'appel

  {"name": "search_blog", "arguments": {"n": 1, "tag": "texte", "sort": "relevance", "query": "texte"}}

  Exemple de retour

  {"content": [{"type": "text", "text": "…"}], "structuredContent": {"result": []}}

C'est votre serveur ?

Revendiquez-le sur l'espace créateur (preuve GitHub, domaine ou npm) pour afficher le badge « propriétaire vérifié » et lancer un pentest comportemental à la demande afin d'obtenir le badge « Vérifié ».

Créer un compte / revendiquer →

Source

https://mcp.sovgrid.org/self-hosted-ai?ref=mcp-io