io.github.aikuba/llmconfigurator

Est-ce que io.github.aikuba/llmconfigurator est sûr ?

Au dernier audit MCP Hub, le serveur MCP « io.github.aikuba/llmconfigurator » 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

• check_hardware_fit≈ 1609 tokens

  Use this when the user asks whether one specific model runs on one specific machine, at what memory cost, and roughly how fast. Needs a hardware id and a model id from search_catalog. Returns a fit verdict, the memory arithmetic, and a decode-speed range with a confidence label. Memory is sized from all parameters, speed from active parameters. It does not recommend purchases and does not run anything.

  Paramètres, retour & exemple (4)

  Paramètres

  • model_id · string · requis
    Model id from search_catalog.
  • hardware_id · string · requis
    Hardware id from search_catalog.
  • quantisation · string · optionnel · Q2_K | Q3_K_M | Q4_K_M | Q5_K_M | Q6_K | Q8_0 | F16
    Weight quantisation. One of Q2_K, Q3_K_M, Q4_K_M, Q5_K_M, Q6_K, Q8_0, F16. Default Q4_K_M.
  • context_length · integer · optionnel
    Context window in tokens that the KV cache is sized for. 512 to 262144. Default 4096.

  Exemple d'appel

  {"name": "check_hardware_fit", "arguments": {"model_id": "texte", "hardware_id": "texte"}}

  Exemple de retour

  {"content": [{"type": "text", "text": "…"}], "structuredContent": {"url": "texte", "model": {}, "notes": [], "speed": "texte", "memory": {}, "sources": [], "summary": "texte", "verdict": "fits_comfortably", "hardware": {}, "data_as_of": "texte", "quantisation": "Q2_K", "context_length": 1}}

• compare_hardware≈ 842 tokens

  Use this when the user compares two to four machines. Needs hardware ids from search_catalog. Returns memory, bandwidth, memory type, release year, status and a dated price where one is recorded, side by side. A price of null means none is on record. It does not check any model against the machines; use check_hardware_fit for that.

  Paramètres, retour & exemple (1)

  Paramètres

  • hardware_ids · array · requis
    Two to four hardware ids from search_catalog.

  Exemple d'appel

  {"name": "compare_hardware", "arguments": {"hardware_ids": []}}

  Exemple de retour

  {"content": [{"type": "text", "text": "…"}], "structuredContent": {"summary": "texte", "hardware": [], "data_as_of": "texte"}}

• get_model_specs≈ 796 tokens

  Use this when the user asks about one model's size, architecture, context window or licence. Needs a model id from search_catalog. Returns total and active parameters, weight size per quantisation, context length, licence, release date and sources. It does not say whether the model fits any machine; use check_hardware_fit for that.

  Paramètres, retour & exemple (1)

  Paramètres

  • model_id · string · requis
    Model id from search_catalog.

  Exemple d'appel

  {"name": "get_model_specs", "arguments": {"model_id": "texte"}}

  Exemple de retour

  {"content": [{"type": "text", "text": "…"}], "structuredContent": {"url": "texte", "model": {}, "sources": [], "summary": "texte", "data_as_of": "texte", "quantisations": []}}

• models_for_hardware≈ 1222 tokens

  Use this when the user asks what models a given machine can run. Needs a hardware id from search_catalog. Returns up to 20 catalogue models ranked for that machine, each with a fit verdict and a speed range, optionally filtered to one use such as coding. It covers only models in the catalogue, at one quantisation and context length per call, and leaves out models that do not fit.

  Paramètres, retour & exemple (5)

  Paramètres

  • limit · integer · optionnel
    Maximum models, 1 to 20. Default 10.
  • use_case · string · optionnel · coding | chat | reasoning | agents | vision
    Only list models tagged for this use. One of coding, chat, reasoning, agents, vision. Omit for all models.
  • hardware_id · string · requis
    Hardware id from search_catalog.
  • quantisation · string · optionnel · Q2_K | Q3_K_M | Q4_K_M | Q5_K_M | Q6_K | Q8_0 | F16
    Weight quantisation. One of Q2_K, Q3_K_M, Q4_K_M, Q5_K_M, Q6_K, Q8_0, F16. Default Q4_K_M.
  • context_length · integer · optionnel
    Context window in tokens that the KV cache is sized for. 512 to 262144. Default 4096.

  Exemple d'appel

  {"name": "models_for_hardware", "arguments": {"hardware_id": "texte"}}

  Exemple de retour

  {"content": [{"type": "text", "text": "…"}], "structuredContent": {"url": "texte", "models": [], "summary": "texte", "hardware": {}, "use_case": "texte", "data_as_of": "texte", "quantisation": "Q2_K", "context_length": 1}}

• search_catalog≈ 530 tokens

  Use this first when the user names a GPU, Mac, mini PC or language model and you need its id for the other tools. Matches names and common aliases in a fixed catalogue of local-LLM hardware and models. It does not return specs, fit or speed, and it does not search the web.

  Paramètres, retour & exemple (3)

  Paramètres

  • kind · string · requis · hardware | model
    Search hardware or models. Required.
  • limit · integer · optionnel
    Maximum matches, 1 to 10. Default 5.
  • query · string · requis
    Hardware or model name as the user wrote it, e.g. "4090", "m4 max 64", "llama 3.3 70b".

  Exemple d'appel

  {"name": "search_catalog", "arguments": {"kind": "hardware", "query": "texte"}}

  Exemple de retour

  {"content": [{"type": "text", "text": "…"}], "structuredContent": {"kind": "hardware", "matches": [], "summary": "texte", "data_as_of": "texte"}}

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://elgadmisijrfowuujcmu.supabase.co/functions/v1/mcp