io.github.insivotron/agent-utility-mcp

Est-ce que io.github.insivotron/agent-utility-mcp est sûr ?

Au dernier audit MCP Hub, le serveur MCP « io.github.insivotron/agent-utility-mcp » 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 4 outils

• analyze_tool_call≈ 1029 tokens

  Analyze a proposed AI-agent tool call before execution and return a deterministic, policy-aware decision (allow, notice, confirm or block) covering destructive actions, sensitive-data exposure, external transmission, privilege changes and irreversible operations, under the applicable policy — permissive, balanced or strict, balanced by default. Filesystem paths outside a known workspace_root are treated as higher risk than paths inside it.

  Paramètres, retour & exemple (4)

  Paramètres

  • policy · string · optionnel · permissive | balanced | strict
    Optional policy used to compute the decision: permissive, balanced or strict (default: balanced).
  • context · object · optionnel
    Optional declarative context about the proposed operation.
  • arguments · ? · requis
    The proposed tool arguments. They are analyzed as data and never executed.
  • tool_name · string · requis
    The exact name of the proposed tool.

  Exemple d'appel

  {"name": "analyze_tool_call", "arguments": {"arguments": "texte", "tool_name": "texte"}}

  Exemple de retour

  {"content": [{"type": "text", "text": "…"}], "structuredContent": {"risk": "safe", "valid": true, "checks": {}, "signals": [], "decision": "allow", "tool_name": "texte", "risk_score": 1, "policy_applied": "permissive", "schema_version": "texte", "ruleset_version": "texte", "context_completeness": "full"}}

• analyze_url_risk≈ 874 tokens

  Analyze a URL for structural and security risk signals and return a deterministic, policy-aware decision (allow, notice, confirm or block) under the applicable policy — permissive, balanced or strict, balanced by default.

  Paramètres, retour & exemple (3)

  Paramètres

  • url · string · requis
    The absolute URL to analyze.
  • policy · string · optionnel · permissive | balanced | strict
    Optional policy used to compute the decision: permissive, balanced or strict (default: balanced).
  • context · object · optionnel
    Optional declarative context.

  Exemple d'appel

  {"name": "analyze_url_risk", "arguments": {"url": "texte"}}

  Exemple de retour

  {"content": [{"type": "text", "text": "…"}], "structuredContent": {"port": "texte", "risk": "safe", "input": "texte", "valid": true, "checks": {}, "domain": "texte", "signals": [], "decision": "allow", "hostname": "texte", "protocol": "texte", "subdomain": "texte", "risk_score": 1}}

• inspect_command≈ 973 tokens

  Analyze a shell command before execution and return a deterministic, policy-aware decision (allow, notice, confirm or block) without running it, under the applicable policy — permissive, balanced or strict, balanced by default. Paths outside a known workspace_root are treated as higher risk than paths inside it.

  Paramètres, retour & exemple (5)

  Paramètres

  • cwd · string · optionnel
    Optional working-directory context. It is never accessed.
  • shell · string · optionnel · bash | sh | zsh | powershell | cmd | unknown
    The command shell, when known.
  • policy · string · optionnel · permissive | balanced | strict
    Optional policy used to compute the decision: permissive, balanced or strict (default: balanced).
  • command · string · requis
    The complete command text to inspect without executing it.
  • context · object · optionnel
    Optional declarative context.

  Exemple d'appel

  {"name": "inspect_command", "arguments": {"command": "texte"}}

  Exemple de retour

  {"content": [{"type": "text", "text": "…"}], "structuredContent": {"cwd": "texte", "risk": "safe", "shell": "bash", "valid": true, "checks": {}, "command": "texte", "signals": [], "decision": "allow", "risk_score": 1, "policy_applied": "permissive", "schema_version": "texte", "ruleset_version": "texte"}}

• inspect_file≈ 820 tokens

  Inspect a Base64-encoded file locally for deterministic structural and security risk signals without executing it, and return a policy-aware decision (allow, notice, confirm or block) under the applicable policy — permissive, balanced or strict, balanced by default.

  Paramètres, retour & exemple (4)

  Paramètres

  • policy · string · optionnel · permissive | balanced | strict
    Optional policy used to compute the decision: permissive, balanced or strict (default: balanced).
  • context · object · optionnel
    Optional declarative context.
  • filename · string · requis
    The original filename, including its extension.
  • content_base64 · string · requis
    The complete file content encoded as canonical Base64 (maximum decoded size: 1 MiB).

  Exemple d'appel

  {"name": "inspect_file", "arguments": {"filename": "texte", "content_base64": "texte"}}

  Exemple de retour

  {"content": [{"type": "text", "text": "…"}], "structuredContent": {"risk": "safe", "valid": true, "checks": {}, "sha256": "texte", "signals": [], "decision": "allow", "filename": "texte", "risk_score": 1, "size_bytes": 1, "detected_type": "texte", "policy_applied": "permissive", "schema_version": "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://agent-utility-mcp.insivotron.workers.dev/mcp