Koskamo

Agent Utility MCP

A deterministic, read-only security preflight layer for AI agents that analyzes shell commands, tool calls, URLs, and files for destructive or high-risk behavior without executing them or making external API calls.

Endpoint and access

No sign-in to connect and list tools. Individual tools may still need a vendor account, credentials, or permissions; check the documented prerequisites.

Agent Utility MCP logo, by InsivotronCommunity listingStreamable HTTPNo sign-inEndpoint checked 2026-11-06Published by Insivotron
MCP endpoint (Streamable HTTP)
https://agent-utility-mcp.insivotron.workers.dev/mcp
Vendor documentation

What its tools do

4 tools, grouped by the hints the server advertises. Hints are self-declared, not verified behavior.

Read-only (4)

4 of 4 have a description · 4 of 4 declare behavior hints · 4 of 4 with parameters document them all

Checks and signals

Each line is separate evidence, not a trust score.

  • Passed Endpoint answered the live check ·
  • Availability 1 of 1 checks passed in 30 days

What you can do with it

  • Analyze URLs for structural and security risk signals before access
  • Inspect file metadata and content signatures without executing the file
  • Analyze shell commands for destructive, privileged, or remote-execution behavior
  • Detect destructive actions and sensitive-data exposure in proposed agent tool calls

Source and maintenance

Community listingRegistry 0.3.0
mcp-registry: io.github.insivotron/agent-utility-mcp@0.3.0

Protocol and capabilities

Server
agent-utility-mcp-v1 0.3.0
MCP protocol negotiated in check
2025-11-25
Tools
4 advertised
Resources
Not advertised
Resource templates
Not advertised
Prompts
Not advertised
Server-advertised capabilities
{
  "tools": {
    "listChanged": true
  }
}

Tools

4 advertised

analyze_url_risk
Advertised by the server during the last successful anonymous check.
Advertised read-only

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.

analyze_url_risk input schema
{
  "type": "object",
  "$schema": "http://json-schema.org/draft-07/schema#",
  "required": [
    "url"
  ],
  "properties": {
    "url": {
      "type": "string",
      "minLength": 1,
      "description": "The absolute URL to analyze."
    },
    "policy": {
      "enum": [
        "permissive",
        "balanced",
        "strict"
      ],
      "type": "string",
      "description": "Optional policy used to compute the decision: permissive, balanced or strict (default: balanced)."
    },
    "context": {
      "type": "object",
      "properties": {
        "workspace_root": {
          "type": "string",
          "description": "Accepted for contract consistency; URL analysis has no filesystem paths to scope, so it has no effect here."
        }
      },
      "description": "Optional declarative context.",
      "additionalProperties": false
    }
  },
  "additionalProperties": false
}
analyze_url_risk output schema
{
  "type": "object",
  "$schema": "http://json-schema.org/draft-07/schema#",
  "required": [
    "valid",
    "input",
    "normalized_url",
    "hostname",
    "domain",
    "public_suffix",
    "subdomain",
    "protocol",
    "port",
    "risk",
    "risk_score",
    "decision",
    "policy_applied",
    "ruleset_version",
    "schema_version",
    "context_completeness",
    "signals",
    "checks"
  ],
  "properties": {
    "port": {
      "type": [
        "string",
        "null"
      ]
    },
    "risk": {
      "enum": [
        "safe",
        "low",
        "medium",
        "high",
        "critical"
      ],
      "type": "string"
    },
    "input": {
      "type": "string"
    },
    "valid": {
      "type": "boolean"
    },
    "checks": {
      "type": "object",
      "required": [
        "dangerous_scheme",
        "embedded_credentials",
        "raw_ip_host",
        "private_or_local_host",
        "cloud_metadata_host",
        "unicode_or_punycode_host",
        "suspicious_port",
        "excessive_subdomains",
        "suspicious_keywords",
        "executable_path",
        "url_shortener",
        "excessive_length"
      ],
      "properties": {
        "raw_ip_host": {
          "type": "boolean"
        },
        "url_shortener": {
          "type": "boolean"
        },
        "executable_path": {
          "type": "boolean"
        },
        "suspicious_port": {
          "type": "boolean"
        },
        "dangerous_scheme": {
          "type": "boolean"
        },
        "excessive_length": {
          "type": "boolean"
        },
        "cloud_metadata_host": {
          "type": "boolean"
        },
        "suspicious_keywords": {
          "type": "boolean"
        },
        "embedded_credentials": {
          "type": "boolean"
        },
        "excessive_subdomains": {
          "type": "boolean"
        },
        "private_or_local_host": {
          "type": "boolean"
        },
        "unicode_or_punycode_host": {
          "type": "boolean"
        }
      },
      "additionalProperties": false
    },
    "domain": {
      "type": [
        "string",
        "null"
      ]
    },
    "signals": {
      "type": "array",
      "items": {
        "type": "object",
        "required": [
          "code",
          "severity",
          "score",
          "message"
        ],
        "properties": {
          "code": {
            "type": "string"
          },
          "score": {
            "type": "number"
          },
          "message": {
            "type": "string"
          },
          "severity": {
            "enum": [
              "info",
              "low",
              "medium",
              "high",
              "critical"
            ],
            "type": "string"
          }
        },
        "additionalProperties": false
      }
    },
    "decision": {
      "enum": [
        "allow",
        "notice",
        "confirm",
        "block"
      ],
      "type": "string"
    },
    "hostname": {
      "type": [
        "string",
        "null"
      ]
    },
    "protocol": {
      "type": [
        "string",
        "null"
      ]
    },
    "subdomain": {
      "type": [
        "string",
        "null"
      ]
    },
    "risk_score": {
      "type": "number"
    },
    "public_suffix": {
      "type": [
        "string",
        "null"
      ]
    },
    "normalized_url": {
      "type": [
        "string",
        "null"
      ]
    },
    "policy_applied": {
      "enum": [
        "permissive",
        "balanced",
        "strict"
      ],
      "type": "string"
    },
    "schema_version": {
      "type": "string"
    },
    "ruleset_version": {
      "type": "string"
    },
    "context_completeness": {
      "enum": [
        "full",
        "partial",
        "none"
      ],
      "type": "string"
    }
  },
  "additionalProperties": false
}
analyze_url_risk annotations
{
  "readOnlyHint": true,
  "openWorldHint": false,
  "idempotentHint": true,
  "destructiveHint": false
}
inspect_command
Advertised by the server during the last successful anonymous check.
Advertised read-only

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.

inspect_command input schema
{
  "type": "object",
  "$schema": "http://json-schema.org/draft-07/schema#",
  "required": [
    "command"
  ],
  "properties": {
    "cwd": {
      "type": "string",
      "maxLength": 4096,
      "description": "Optional working-directory context. It is never accessed."
    },
    "shell": {
      "enum": [
        "bash",
        "sh",
        "zsh",
        "powershell",
        "cmd",
        "unknown"
      ],
      "type": "string",
      "description": "The command shell, when known."
    },
    "policy": {
      "enum": [
        "permissive",
        "balanced",
        "strict"
      ],
      "type": "string",
      "description": "Optional policy used to compute the decision: permissive, balanced or strict (default: balanced)."
    },
    "command": {
      "type": "string",
      "maxLength": 16384,
      "minLength": 1,
      "description": "The complete command text to inspect without executing it."
    },
    "context": {
      "type": "object",
      "properties": {
        "workspace_root": {
          "type": "string",
          "maxLength": 4096,
          "description": "Optional workspace root. Paths that resolve outside it are treated as higher risk than paths inside it."
        }
      },
      "description": "Optional declarative context.",
      "additionalProperties": false
    }
  },
  "additionalProperties": false
}
inspect_command output schema
{
  "type": "object",
  "$schema": "http://json-schema.org/draft-07/schema#",
  "required": [
    "valid",
    "command",
    "shell",
    "cwd",
    "risk",
    "risk_score",
    "decision",
    "policy_applied",
    "ruleset_version",
    "schema_version",
    "context_completeness",
    "signals",
    "checks"
  ],
  "properties": {
    "cwd": {
      "type": [
        "string",
        "null"
      ]
    },
    "risk": {
      "enum": [
        "safe",
        "low",
        "medium",
        "high",
        "critical"
      ],
      "type": "string"
    },
    "shell": {
      "enum": [
        "bash",
        "sh",
        "zsh",
        "powershell",
        "cmd",
        "unknown"
      ],
      "type": "string"
    },
    "valid": {
      "type": "boolean"
    },
    "checks": {
      "type": "object",
      "required": [
        "destructive_filesystem",
        "root_target",
        "privilege_escalation",
        "network_access",
        "network_download",
        "network_transfer",
        "remote_code_execution",
        "sensitive_file_access",
        "possible_exfiltration",
        "permission_change",
        "persistence_change",
        "system_modification",
        "obfuscated_execution",
        "command_chaining",
        "shell_spawning"
      ],
      "properties": {
        "root_target": {
          "type": "boolean"
        },
        "network_access": {
          "type": "boolean"
        },
        "shell_spawning": {
          "type": "boolean"
        },
        "command_chaining": {
          "type": "boolean"
        },
        "network_download": {
          "type": "boolean"
        },
        "network_transfer": {
          "type": "boolean"
        },
        "permission_change": {
          "type": "boolean"
        },
        "persistence_change": {
          "type": "boolean"
        },
        "system_modification": {
          "type": "boolean"
        },
        "obfuscated_execution": {
          "type": "boolean"
        },
        "privilege_escalation": {
          "type": "boolean"
        },
        "possible_exfiltration": {
          "type": "boolean"
        },
        "remote_code_execution": {
          "type": "boolean"
        },
        "sensitive_file_access": {
          "type": "boolean"
        },
        "destructive_filesystem": {
          "type": "boolean"
        }
      },
      "additionalProperties": false
    },
    "command": {
      "type": [
        "string",
        "null"
      ]
    },
    "signals": {
      "type": "array",
      "items": {
        "type": "object",
        "required": [
          "code",
          "severity",
          "score",
          "message"
        ],
        "properties": {
          "code": {
            "type": "string"
          },
          "score": {
            "type": "number"
          },
          "message": {
            "type": "string"
          },
          "severity": {
            "enum": [
              "info",
              "low",
              "medium",
              "high",
              "critical"
            ],
            "type": "string"
          }
        },
        "additionalProperties": false
      }
    },
    "decision": {
      "enum": [
        "allow",
        "notice",
        "confirm",
        "block"
      ],
      "type": "string"
    },
    "risk_score": {
      "type": "number"
    },
    "policy_applied": {
      "enum": [
        "permissive",
        "balanced",
        "strict"
      ],
      "type": "string"
    },
    "schema_version": {
      "type": "string"
    },
    "ruleset_version": {
      "type": "string"
    },
    "context_completeness": {
      "enum": [
        "full",
        "partial",
        "none"
      ],
      "type": "string"
    }
  },
  "additionalProperties": false
}
inspect_command annotations
{
  "readOnlyHint": true,
  "openWorldHint": false,
  "idempotentHint": true,
  "destructiveHint": false
}
inspect_file
Advertised by the server during the last successful anonymous check.
Advertised read-only

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.

inspect_file input schema
{
  "type": "object",
  "$schema": "http://json-schema.org/draft-07/schema#",
  "required": [
    "filename",
    "content_base64"
  ],
  "properties": {
    "policy": {
      "enum": [
        "permissive",
        "balanced",
        "strict"
      ],
      "type": "string",
      "description": "Optional policy used to compute the decision: permissive, balanced or strict (default: balanced)."
    },
    "context": {
      "type": "object",
      "properties": {
        "workspace_root": {
          "type": "string",
          "description": "Accepted for contract consistency; file inspection has no filesystem paths to scope, so it has no effect here."
        }
      },
      "description": "Optional declarative context.",
      "additionalProperties": false
    },
    "filename": {
      "type": "string",
      "minLength": 1,
      "description": "The original filename, including its extension."
    },
    "content_base64": {
      "type": "string",
      "description": "The complete file content encoded as canonical Base64 (maximum decoded size: 1 MiB)."
    }
  },
  "additionalProperties": false
}
inspect_file output schema
{
  "type": "object",
  "$schema": "http://json-schema.org/draft-07/schema#",
  "required": [
    "valid",
    "filename",
    "detected_type",
    "declared_extension",
    "size_bytes",
    "sha256",
    "risk",
    "risk_score",
    "decision",
    "policy_applied",
    "ruleset_version",
    "schema_version",
    "context_completeness",
    "signals",
    "checks"
  ],
  "properties": {
    "risk": {
      "enum": [
        "safe",
        "low",
        "medium",
        "high",
        "critical"
      ],
      "type": "string"
    },
    "valid": {
      "type": "boolean"
    },
    "checks": {
      "type": "object",
      "required": [
        "empty_file",
        "exceeds_size_limit",
        "dangerous_extension",
        "double_extension",
        "executable_content",
        "extension_type_mismatch",
        "suspicious_filename"
      ],
      "properties": {
        "empty_file": {
          "type": "boolean"
        },
        "double_extension": {
          "type": "boolean"
        },
        "exceeds_size_limit": {
          "type": "boolean"
        },
        "executable_content": {
          "type": "boolean"
        },
        "dangerous_extension": {
          "type": "boolean"
        },
        "suspicious_filename": {
          "type": "boolean"
        },
        "extension_type_mismatch": {
          "type": "boolean"
        }
      },
      "additionalProperties": false
    },
    "sha256": {
      "type": [
        "string",
        "null"
      ]
    },
    "signals": {
      "type": "array",
      "items": {
        "type": "object",
        "required": [
          "code",
          "severity",
          "score",
          "message"
        ],
        "properties": {
          "code": {
            "type": "string"
          },
          "score": {
            "type": "number"
          },
          "message": {
            "type": "string"
          },
          "severity": {
            "enum": [
              "info",
              "low",
              "medium",
              "high",
              "critical"
            ],
            "type": "string"
          }
        },
        "additionalProperties": false
      }
    },
    "decision": {
      "enum": [
        "allow",
        "notice",
        "confirm",
        "block"
      ],
      "type": "string"
    },
    "filename": {
      "type": [
        "string",
        "null"
      ]
    },
    "risk_score": {
      "type": "number"
    },
    "size_bytes": {
      "type": [
        "number",
        "null"
      ]
    },
    "detected_type": {
      "type": [
        "string",
        "null"
      ]
    },
    "policy_applied": {
      "enum": [
        "permissive",
        "balanced",
        "strict"
      ],
      "type": "string"
    },
    "schema_version": {
      "type": "string"
    },
    "ruleset_version": {
      "type": "string"
    },
    "declared_extension": {
      "type": [
        "string",
        "null"
      ]
    },
    "context_completeness": {
      "enum": [
        "full",
        "partial",
        "none"
      ],
      "type": "string"
    }
  },
  "additionalProperties": false
}
inspect_file annotations
{
  "readOnlyHint": true,
  "openWorldHint": false,
  "idempotentHint": true,
  "destructiveHint": false
}
analyze_tool_call
Advertised by the server during the last successful anonymous check.
Advertised read-only

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.

analyze_tool_call input schema
{
  "type": "object",
  "$schema": "http://json-schema.org/draft-07/schema#",
  "required": [
    "tool_name",
    "arguments"
  ],
  "properties": {
    "policy": {
      "enum": [
        "permissive",
        "balanced",
        "strict"
      ],
      "type": "string",
      "description": "Optional policy used to compute the decision: permissive, balanced or strict (default: balanced)."
    },
    "context": {
      "type": "object",
      "properties": {
        "operation": {
          "type": "string",
          "maxLength": 4096
        },
        "description": {
          "type": "string",
          "maxLength": 4096
        },
        "destination": {
          "type": "string",
          "maxLength": 4096
        },
        "target_type": {
          "type": "string",
          "maxLength": 4096
        },
        "workspace_root": {
          "type": "string",
          "maxLength": 4096,
          "description": "Optional workspace root. Filesystem paths that resolve outside it are treated as higher risk than paths inside it."
        }
      },
      "description": "Optional declarative context about the proposed operation.",
      "additionalProperties": false
    },
    "arguments": {
      "anyOf": [
        {
          "type": "object",
          "additionalProperties": {}
        },
        {
          "type": "array",
          "items": {}
        },
        {
          "type": "string"
        },
        {
          "type": "number"
        },
        {
          "type": "boolean"
        },
        {
          "type": "null"
        }
      ],
      "description": "The proposed tool arguments. They are analyzed as data and never executed."
    },
    "tool_name": {
      "type": "string",
      "maxLength": 256,
      "minLength": 1,
      "description": "The exact name of the proposed tool."
    }
  },
  "additionalProperties": false
}
analyze_tool_call output schema
{
  "type": "object",
  "$schema": "http://json-schema.org/draft-07/schema#",
  "required": [
    "valid",
    "tool_name",
    "risk",
    "risk_score",
    "decision",
    "policy_applied",
    "ruleset_version",
    "schema_version",
    "context_completeness",
    "signals",
    "checks"
  ],
  "properties": {
    "risk": {
      "enum": [
        "safe",
        "low",
        "medium",
        "high",
        "critical"
      ],
      "type": "string"
    },
    "valid": {
      "type": "boolean"
    },
    "checks": {
      "type": "object",
      "required": [
        "destructive_action",
        "irreversible_action",
        "privilege_change",
        "permission_change",
        "sensitive_data_present",
        "possible_secret_exposure",
        "external_destination",
        "possible_exfiltration",
        "financial_action",
        "communication_action",
        "account_or_identity_action",
        "code_execution",
        "filesystem_mutation",
        "network_action"
      ],
      "properties": {
        "code_execution": {
          "type": "boolean"
        },
        "network_action": {
          "type": "boolean"
        },
        "financial_action": {
          "type": "boolean"
        },
        "privilege_change": {
          "type": "boolean"
        },
        "permission_change": {
          "type": "boolean"
        },
        "destructive_action": {
          "type": "boolean"
        },
        "filesystem_mutation": {
          "type": "boolean"
        },
        "irreversible_action": {
          "type": "boolean"
        },
        "communication_action": {
          "type": "boolean"
        },
        "external_destination": {
          "type": "boolean"
        },
        "possible_exfiltration": {
          "type": "boolean"
        },
        "sensitive_data_present": {
          "type": "boolean"
        },
        "possible_secret_exposure": {
          "type": "boolean"
        },
        "account_or_identity_action": {
          "type": "boolean"
        }
      },
      "additionalProperties": false
    },
    "signals": {
      "type": "array",
      "items": {
        "type": "object",
        "required": [
          "code",
          "severity",
          "score",
          "message"
        ],
        "properties": {
          "code": {
            "type": "string"
          },
          "score": {
            "type": "number"
          },
          "message": {
            "type": "string"
          },
          "severity": {
            "enum": [
              "info",
              "low",
              "medium",
              "high",
              "critical"
            ],
            "type": "string"
          }
        },
        "additionalProperties": false
      }
    },
    "decision": {
      "enum": [
        "allow",
        "notice",
        "confirm",
        "block"
      ],
      "type": "string"
    },
    "tool_name": {
      "type": [
        "string",
        "null"
      ]
    },
    "risk_score": {
      "type": "number"
    },
    "policy_applied": {
      "enum": [
        "permissive",
        "balanced",
        "strict"
      ],
      "type": "string"
    },
    "schema_version": {
      "type": "string"
    },
    "ruleset_version": {
      "type": "string"
    },
    "context_completeness": {
      "enum": [
        "full",
        "partial",
        "none"
      ],
      "type": "string"
    }
  },
  "additionalProperties": false
}
analyze_tool_call annotations
{
  "readOnlyHint": true,
  "openWorldHint": false,
  "idempotentHint": true,
  "destructiveHint": false
}

Prompts

Not advertised

No prompt definitions available.

Resources

Not advertised

No resource definitions available.

90-day uptime

A lightweight connection check about every six hours, sampled on hosts with many listings. It shows the server answered, not that its tools work.

Too few checks for a percentage yet (2 of 2 answered).

  1. 2026-11-06: 1 of 1 checks answered
  2. 2026-11-07: 1 of 1 checks answered

30-day availability

Aggregated from read-only remote checks; this is not a tool-execution or security guarantee.

Successful checks
1 of 1 (100%)
Average latency
396 ms
95th percentile latency
396 ms

Maintenance history

Public check and source-refresh events. Editorial and ownership records remain private.

  • · probe · passed · 396 ms · metadata changed · 4 tool(s)

Connect from your client

Each snippet registers the same endpoint; pick the client you use.

This server is operated by the named third party. Listing and endpoint checks are not an endorsement or a security audit. Review the operator's permissions, terms, and data handling before connecting. How listings are checked and corrected.

Claude Code

Run this once in a terminal. It registers the remote server. Use --scope project to write .mcp.json for the repo.
Claude Code configuration
claude mcp add --transport http agent-utility-mcp https://agent-utility-mcp.insivotron.workers.dev/mcp

Codex

Run this once. Codex stores MCP servers in ~/.codex/config.toml (or .codex/config.toml in a trusted project).
Codex configuration
codex mcp add agent-utility-mcp --url https://agent-utility-mcp.insivotron.workers.dev/mcp

Cursor

Merge this server into .cursor/mcp.json in your project, then enable agent-utility-mcp in your MCP settings.
Cursor configuration
{
  "mcpServers": {
    "agent-utility-mcp": {
      "url": "https://agent-utility-mcp.insivotron.workers.dev/mcp"
    }
  }
}

Cline

Merge this into cline_mcp_settings.json (Cline → MCP Servers → Configure).
Cline configuration
{
  "mcpServers": {
    "agent-utility-mcp": {
      "type": "streamableHttp",
      "url": "https://agent-utility-mcp.insivotron.workers.dev/mcp",
      "disabled": false
    }
  }
}

VS Code

Merge this server into .vscode/mcp.json. Use MCP: List Servers to start agent-utility-mcp and inspect its tools.
VS Code configuration
{
  "servers": {
    "agent-utility-mcp": {
      "type": "http",
      "url": "https://agent-utility-mcp.insivotron.workers.dev/mcp"
    }
  }
}

Gemini CLI

Run this once. It writes .gemini/settings.json in the current project; pass --scope user for ~/.gemini/settings.json.
Gemini CLI configuration
gemini mcp add --transport http agent-utility-mcp https://agent-utility-mcp.insivotron.workers.dev/mcp

Claude Desktop

Open Claude Settings > Connectors > Add custom connector, enter this server URL and enable it for the conversation. This is a manual remote-connector handoff, not a desktop JSON configuration.
Claude Desktop configuration
https://agent-utility-mcp.insivotron.workers.dev/mcp

Windsurf

Merge this into ~/.codeium/windsurf/mcp_config.json, then refresh MCP in Cascade. Windsurf remote servers use serverUrl, not url.
Windsurf configuration
{
  "mcpServers": {
    "agent-utility-mcp": {
      "serverUrl": "https://agent-utility-mcp.insivotron.workers.dev/mcp"
    }
  }
}

Let your agent find its tools

Find servers like Agent Utility MCP from your agent. Add Koskamo once to search the index for endpoints, access requirements, and client setup. Public discovery needs no key.

Connect Koskamo Explore workflows
See all 8,336