Plug and Ship

Search agent tools

search_tools

Read-onlyIdempotent

Find MCP servers, agent skills (SKILL.md) and coding-agent plugins for a task. Pass a plain-language task (e.g. 'query my Postgres from Cursor', 'post to Slack from Claude Code') or keywords. The task is parsed for kind (MCP/skill/plugin), client (Cursor, Claude Code, Codex, VS Code), remote vs local and language; results are ranked by relevance (full-text + semantic match), with stars, recent commits, a valid one-click config and a license only as tie-breakers. Every result explains why it matches. type, client and remote are strict filters (client grok-bot = MCP servers with a public HTTPS endpoint).

Input Schema

TableJSON Schema

Name	Required	Description
`task`	No	What you want to do, in plain language. Used instead of query when given.
`type`	No	Only this kind of tool
`limit`	No	Max results, default 10
`query`	No	Keywords or a short task, e.g. 'postgres', 'browser automation', 'query my Postgres from Cursor'
`client`	No	Only tools that name this coding agent (grok-bot: hosted HTTPS MCP servers)
`remote`	No	Only MCP servers with a public HTTPS endpoint (nothing to install)

Output Schema

TableJSON Schema

Name	Required	Description	Default
`count`	Yes
`query`	Yes
`results`	Yes
`interpreted`	Yes

Schema Changelog

Changes observed during successful MCP inspections.

Changed7 schema fields changedOct 16, 2026

changedInput schema / properties / client / description

Previous value: -"Only tools for this coding agent"New value: +"Only tools that name this coding agent (grok-bot: hosted HTTPS MCP servers)"

changedInput schema / properties / query / description

Previous value: -"Keywords, e.g. 'postgres', 'figma', 'browser automation'"New value: +"Keywords or a short task, e.g. 'postgres', 'browser automation', 'query my Postgres from Cursor'"

changedInput schema / properties / query / maxLength
```
Previous value: -100New value: +300
```

addedInput schema / properties / remote

Added value: +{
+  "description": "Only MCP servers with a public HTTPS endpoint (nothing to install)",
+  "type": "boolean"
+}

addedInput schema / properties / task

Added value: +{
+  "description": "What you want to do, in plain language. Used instead of query when given.",
+  "maxLength": 300,
+  "type": "string"
+}

removedInput schema / required
```
Removed value: -[
-  "query"
-]
```

changedOutput schema / (root)

Previous value: -nullNew value: +{
+  "$schema": "http://json-schema.org/draft-07/schema#",
+  "additionalProperties": false,
+  "properties": {
+    "count": {
+      "type": "number"
+    },
+    "interpreted": {
+      "anyOf": [
+        {
+          "additionalProperties": false,
+          "properties": {
+            "client": {
+              "type": [
+                "string",
+                "null"
+              ]
+            },
+            "kind": {
+              "type": [
+                "string",
+                "null"
+              ]
+            },
+            "language": {
+              "type": [
+                "string",
+                "null"
+              ]
+            },
+            "local": {
+              "type": "boolean"
+            },
+            "remote": {
+              "type": "boolean"
+            },
+            "terms": {
+              "items": {
+                "type": "string"
+              },
+              "type": "array"
+            }
+          },
+          "required": [
+            "kind",
+            "client",
+            "remote",
+            "local",
+            "language",
+            "terms"
+          ],
+          "type": "object"
+        },
+        {
+          "type": "null"
+        }
+      ]
+    },
+    "query": {
+      "type": "string"
+    },
+    "results": {
+      "items": {
+        "additionalProperties": false,
+        "properties": {
+          "description": {
+            "type": [
+              "string",
+              "null"
+            ]
+          },
+          "install": {
+            "type": [
+              "string",
+              "null"
+            ]
+          },
+          "language": {
+            "type": [
+              "string",
+              "null"
+            ]
+          },
+          "last_commit": {
+            "type": [
+              "string",
+              "null"
+            ]
+          },
+          "license": {
+            "type": [
+              "string",
+              "null"
+            ]
+          },
+          "name": {
+            "type": "string"
+          },
+          "page": {
+            "type": "string"
+          },
+          "reasons": {
+            "items": {
+              "type": "string"
+            },
+            "type": "array"
+          },
+          "remote_endpoint": {
+            "type": [
+              "string",
+              "null"
+            ]
+          },
+          "score": {
+            "type": "number"
+          },
+          "slug": {
+            "type": "string"
+          },
+          "stars": {
+            "type": "number"
+          },
+          "type": {
+            "enum": [
+              "mcp",
+              "skill",
+              "plugin"
+            ],
+            "type": "string"
+          },
+          "why": {
+            "type": "string"
+          }
+        },
+        "required": [
+          "slug",
+          "name",
+          "type",
+          "description",
+          "stars",
+          "install",
+          "remote_endpoint",
+          "page",
+          "language",
+          "license",
+          "last_commit"
+        ],
+        "type": "object"
+      },
+      "type": "array"
+    }
+  },
+  "required": [
+    "query",
+    "interpreted",
+    "count",
+    "results"
+  ],
+  "type": "object"
+}

First observedOct 15, 2026

TDQS

A4.5/5.0

Behavior4/5

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

Annotations already declare readOnly/openWorld/idempotent, so the safety profile is covered. The description goes further and discloses behavior the annotations cannot: how the task is parsed (kind, client, remote vs local, language), how results are ranked (full-text + semantic), what is only a tie-breaker, and that filters are strict.

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

Conciseness5/5

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

Front-loaded with the purpose, then calling guidance, then ranking behavior, then filter semantics — a logical order. It is dense for four sentences but every clause carries information an agent needs.

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 not be explained; the description still notes that each result explains its own match, which is the useful part. Query semantics, filters, ranking and result shape are all covered for a 6-parameter search tool.

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 100% and includes enum values and the task/query precedence, so the baseline is 3. The description adds real semantics on top: 'type, client and remote are strict filters' and the grok-bot special case meaning hosted HTTPS MCP servers, clarifying filter behavior rather than just restating fields.

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?

States a specific verb and resource set ('Find MCP servers, agent skills (SKILL.md) and coding-agent plugins') and is clearly distinguishable from browse-oriented siblings like list_best and trending. The scope of what is searchable is explicit.

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

Usage Guidelines4/5

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

Gives concrete calling guidance ('Pass a plain-language task ... or keywords') with two worked examples, which sets the context for use. It never names the alternatives (get_tool, list_best, trending) or states when a browse-style tool would be better, so the sibling routing is left to inference.

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

All Plug and Ship tools Inspect

Try in Browser

Shareable, User-Owned Context BackendPeople, applications, and agents read, write, and coordinate together in one context lake. Reach it over MCP, a CLI, a Python SDK, or REST.fulcradynamics.comAd

Glama MCP Gateway

Add one secure layer between your agents and this server.