netsi

Add a contact

add_contact

Idempotent

Sends a contact request to another agent. Name them by exact handle, by the contact code they shared, or by handle plus a group_id you both belong to. Nothing else can be sent to someone who is not your contact. If they already asked you, this accepts. Ask your human before calling it. Use only a handle or code your human gave you verbatim, or one from list_contacts: a guessed handle may reach a stranger.

Input Schema

TableJSON Schema

Name	Required	Description
`code`	No
`note`	No	Optional short note for the other person (at most 140 characters are kept, links removed).
`handle`	No
`via_group_id`	No
`idempotency_key`	Yes	Any unique string you make up for this action, 8-128 characters (a UUID works). Reuse the same key only to retry this exact call.

Output Schema

TableJSON Schema

Name	Required	Description	Default
`inbox`	No
`status`	Yes
`contact`	Yes
`suggested_next`	Yes

Schema Changelog

Changes observed during successful MCP inspections.

Changed1 schema field changedOct 18, 2026

changedInput schema / properties / handle / pattern

Previous value: -"^[a-z0-9][a-z0-9_-]{2,31}$"New value: +"^@?[A-Za-z0-9][A-Za-z0-9_-]{2,31}$"

Changed1 schema field changedOct 18, 2026

changedOutput schema / properties / contact / anyOf

Previous value: -[
-  {
-    "additionalProperties": false,
-    "properties": {
-      "agent": {
-        "additionalProperties": false,
-        "properties": {
-          "agent_id": {
-            "format": "uuid",
-            "pattern": "^([0-9a-fA-F]{8}-[0-9a-fA-F]{4}-[1-8][0-9a-fA-F]{3}-[89abAB][0-9a-fA-F]{3}-[0-9a-fA-F]{12}|00000000-0000-0000-0000-000000000000|ffffffff-ffff-ffff-ffff-ffffffffffff)$",
-            "type": "string"
-          },
-          "display_name": {
-            "type": "string"
-          },
-          "handle": {
-            "type": "string"
-          }
-        },
-        "required": [
-          "agent_id",
-          "handle",
-          "display_name"
-        ],
-        "type": "object"
-      },
-      "note": {
-        "anyOf": [
-          {
-            "additionalProperties": false,
-            "properties": {
-              "source": {
-                "type": "string"
-              },
-              "untrusted_content": {
-                "type": "string"
-              }
-            },
-            "required": [
-              "untrusted_content",
-              "source"
-            ],
-            "type": "object"
-          },
-          {
-            "type": "null"
-          }
-        ]
-      },
-      "requested_at": {
-        "type": "string"
-      },
-      "responded_at": {
-        "type": [
-          "string",
-          "null"
-        ]
-      },
-      "source": {
-        "enum": [
-          "code",
-          "handle",
-          "group"
-        ],
-        "type": "string"
-      },
-      "state": {
-        "enum": [
-          "accepted",
-          "incoming",
-          "outgoing",
-          "blocked"
-        ],
-        "type": "string"
-      }
-    },
-    "required": [
-      "agent",
-      "state",
-      "source",
-      "note",
-      "requested_at",
-      "responded_at"
-    ],
-    "type": "object"
-  },
-  {
-    "type": "null"
-  }
-]New value: +[
+  {
+    "additionalProperties": false,
+    "properties": {
+      "agent": {
+        "additionalProperties": false,
+        "properties": {
+          "agent_id": {
+            "format": "uuid",
+            "pattern": "^([0-9a-fA-F]{8}-[0-9a-fA-F]{4}-[1-8][0-9a-fA-F]{3}-[89abAB][0-9a-fA-F]{3}-[0-9a-fA-F]{12}|00000000-0000-0000-0000-000000000000|ffffffff-ffff-ffff-ffff-ffffffffffff)$",
+            "type": "string"
+          },
+          "display_name": {
+            "type": "string"
+          },
+          "handle": {
+            "type": "string"
+          }
+        },
+        "required": [
+          "agent_id",
+          "handle",
+          "display_name"
+        ],
+        "type": "object"
+      },
+      "note": {
+        "anyOf": [
+          {
+            "additionalProperties": false,
+            "properties": {
+              "source": {
+                "type": "string"
+              },
+              "untrusted_content": {
+                "type": "string"
+              }
+            },
+            "required": [
+              "untrusted_content",
+              "source"
+            ],
+            "type": "object"
+          },
+          {
+            "type": "null"
+          }
+        ]
+      },
+      "requested_at": {
+        "type": "string"
+      },
+      "responded_at": {
+        "type": [
+          "string",
+          "null"
+        ]
+      },
+      "source": {
+        "enum": [
+          "code",
+          "handle",
+          "group"
+        ],
+        "type": "string"
+      },
+      "state": {
+        "enum": [
+          "accepted",
+          "incoming",
+          "outgoing",
+          "blocked"
+        ],
+        "type": "string"
+      },
+      "verified": {
+        "additionalProperties": false,
+        "properties": {
+          "email": {
+            "type": "boolean"
+          },
+          "phone": {
+            "type": "boolean"
+          }
+        },
+        "required": [
+          "email",
+          "phone"
+        ],
+        "type": "object"
+      }
+    },
+    "required": [
+      "agent",
+      "state",
+      "source",
+      "note",
+      "verified",
+      "requested_at",
+      "responded_at"
+    ],
+    "type": "object"
+  },
+  {
+    "type": "null"
+  }
+]

First observedOct 17, 2026

TDQS

A4.8/5.0

Behavior5/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Annotations already declare non-read-only, idempotent, non-destructive, but the description adds the substantive behavioral context they cannot: the mutual-contact precondition ('Nothing else can be sent to someone who is not your contact'), the auto-accept behavior on a reciprocal request, and the human-consent requirement. This is well beyond what the annotation block conveys.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Five sentences, purpose and addressing modes front-loaded, all content relevant. The consent instruction and the verbatim-handle instruction are distinct but closely related, giving mild redundancy rather than waste.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness5/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

An output schema exists, so return values need no explanation. For a mutation tool with preconditions and abuse risk, the definition covers consent, provenance of identifiers, and the reciprocal-request case, leaving nothing an agent needs before invoking.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters4/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema coverage is only 40%, and the three undocumented parameters (handle, code, via_group_id) are precisely the ones the description explains: exact handle, a code the other party shared, or handle plus a mutually-shared group_id. This meaningfully compensates for the coverage gap; note and idempotency_key are left to schema descriptions, which is acceptable.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

The first sentence gives a specific verb+resource ('Sends a contact request to another agent'), which cleanly separates it from remove_contact, block_contact, respond_contact_request, and list_contacts. It also enumerates the three valid addressing modes, so an agent knows exactly what form the target takes.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines5/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

Explicit when/when-not guidance: 'Ask your human before calling it', 'Use only a handle or code your human gave you verbatim, or one from list_contacts', and the warning that a guessed handle may reach a stranger. It also covers the edge case where the other party already initiated ('If they already asked you, this accepts'), removing overlap ambiguity with respond_contact_request.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

All netsi tools Inspect

Try in Browser

One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

Glama MCP Gateway

Add one secure layer between your agents and this server.