Local MCP

LMCP State

lmcp_state

Read-only

Returns a structured snapshot of the LMCP environment: server and tray versions, detected AI client, cloud relay state, TCC permission states (Calendar/Reminders/Contacts), and a compact summary of which services (Mail/Calendar/Contacts/Teams/OneDrive/Reminders/Notes) are reachable. Fast (<500ms), passive — never prompts the user, never opens app windows, never touches the network. Call this when you need to verify the environment is healthy before attempting a tool, or to understand what's installed and accessible. If services.scan_pending is true, the background service scan hasn't finished yet (just after startup) and the per-service running/accounts values are placeholders — do NOT treat them as a real outage; just call the tool you need. Otherwise services.scanned_seconds_ago tells you how many seconds ago that scan ran (cadence ~60s): the per-service values are a snapshot, NOT a live probe. A false/0/not available for a service is advisory only — it can be stale (e.g. the user connected WhatsApp or opened Mail seconds ago) — so never use this tool as a preflight gate to skip or cancel a task; the actual tool call is the source of truth, just attempt it. For reporting failures, use report_problem instead — it captures this same snapshot plus logs and submits to the team.

Input Schema

Name	Required	Description	Default
No arguments

Output Schema

Name	Required	Description
`tcc`	No	TCC permission states. Automation keys (<app>_automation) are one of notDetermined \| authorized \| denied \| notRunning.
`arch`	No
`update`	No
`version`	No	Serving (running) server version.
`built_at`	No	UTC build timestamp (F-041). 'unknown' if unstamped.
`services`	No	Per-domain reachability summary (mail, calendar, contacts, teams, onedrive, slack, …); shape varies by domain. May include `scan_pending: true` right after startup, meaning the per-service running/accounts values are placeholders and not yet authoritative. Once scanned, `scanned_seconds_ago` gives the age (seconds) of that background snapshot and `freshness` restates that a false/0/not-available is advisory, not a live check — never gate a task on it.
`ai_client`	No
`build_sha`	No	Git short SHA of the build (F-041). 'unknown' if unstamped.
`machine_id`	No
`os_version`	No
`web_agents`	No	Web (cloud-relay) clients, counts only: `count` connected (absent when unknown), `relayed_calls_since_launch`, `last_remote_call` {at, tool, ok, ago_s}.
`menu_bar_app`	No	Whether the LMCP menu-bar app is installed. When installed is false, only this MCP server is running (started via npx or the Claude Desktop extension), without the menu-bar app; `install` says how the user can get it.
`tray_version`	No
`last_activity`	No
`skill_capture`	No	Dark-launch counters for repeated-workflow detection (user-generated skills, PR-1). would_fire_signatures = how many times a save-this-workflow nudge WOULD have fired; ring_size = entries in the recent-calls ring. PRIVACY: the ring is IN-MEMORY ONLY — never written to disk, never transmitted, cleared on restart — and argument values are PII-scrubbed on entry, so it holds the SHAPE of a workflow, not its content. No UI acts on it yet; it exists to tune thresholds. Opt out with the skill_capture_enabled config flag.
`always_allowed`	No	Kinds of action the person allowed without asking (`actions`, families such as send_email), whether IT forbids it (`forbidden_by_it`), and how to revoke one (set_always_allowed with allowed: false).
`license_status`	No	trial \| active \| expired
`cloud_token_set`	No
`tunnel_connected`	No
`client_autoconfig`	No	What LMCP's last automatic setup did per AI client (keyed by client id): result (configured, already, failed, skipped or blocked by IT policy), reason_key, at, path. Tell the user which apps were connected and which weren't, and why.
`cloud_data_enabled`	No
`relay_token_other_machine`	No	Present (true) only when this Mac's Cloud Relay token was issued for a different machine id; see relay_token_other_machine_fix.
`relay_token_other_machine_fix`	No
`relay_token_other_machine_detail`	No

Schema Changelog

Changes observed during successful MCP inspections.

Changed3 schema fields changedOct 24, 2026

addedOutput schema / properties / always_allowed

Added value: +{
+  "description": "Kinds of action the person allowed without asking (`actions`, families such as send_email), whether IT forbids it (`forbidden_by_it`), and how to revoke one (set_always_allowed with allowed: false).",
+  "type": "object"
+}

addedOutput schema / properties / client_autoconfig

Added value: +{
+  "description": "What LMCP's last automatic setup did per AI client (keyed by client id): result (configured, already, failed, skipped or blocked by IT policy), reason_key, at, path. Tell the user which apps were connected and which weren't, and why.",
+  "type": "object"
+}

addedOutput schema / properties / menu_bar_app

Added value: +{
+  "description": "Whether the LMCP menu-bar app is installed. When installed is false, only this MCP server is running (started via npx or the Claude Desktop extension), without the menu-bar app; `install` says how the user can get it.",
+  "properties": {
+    "install": {
+      "description": "Present only when installed is false.",
+      "type": "string"
+    },
+    "installed": {
+      "type": "boolean"
+    }
+  },
+  "type": "object"
+}

Changed3 schema fields changedOct 10, 2026

addedOutput schema / properties / relay_token_other_machine

Added value: +{
+  "description": "Present (true) only when this Mac's Cloud Relay token was issued for a different machine id; see relay_token_other_machine_fix.",
+  "type": "boolean"
+}

addedOutput schema / properties / relay_token_other_machine_detail
```
Added value: +{
+  "type": "string"
+}
```
addedOutput schema / properties / relay_token_other_machine_fix
```
Added value: +{
+  "type": "string"
+}
```

Changed7 schema fields changedSep 30, 2026

addedOutput schema / properties / last_activity / properties / last_remote

Added value: +{
+  "description": "Last call relayed from a web client (source cloud), when there was one.",
+  "type": "object"
+}

changedOutput schema / properties / tcc / description

Previous value: -"TCC permission states, e.g. granted | denied | authorized."New value: +"TCC permission states. Automation keys (<app>_automation) are one of notDetermined | authorized | denied | notRunning."

addedOutput schema / properties / tcc / properties / mail_automation_last_known

Added value: +{
+  "description": "Present only when mail_automation is notRunning: \"authorized\" if the grant was confirmed before. Not a live verdict.",
+  "type": "string"
+}

addedOutput schema / properties / tcc / properties / messages_automation
```
Added value: +{
+  "type": "string"
+}
```
addedOutput schema / properties / tcc / properties / notes_automation
```
Added value: +{
+  "type": "string"
+}
```

addedOutput schema / properties / tcc / properties / omnifocus_automation

Added value: +{
+  "description": "Each <app>_automation is present only if the app is installed; <app>_last_known and <app>_note may accompany it.",
+  "type": "string"
+}

addedOutput schema / properties / web_agents

Added value: +{
+  "description": "Web (cloud-relay) clients, counts only: `count` connected (absent when unknown), `relayed_calls_since_launch`, `last_remote_call` {at, tool, ok, ago_s}.",
+  "type": "object"
+}

Changed7 schema fields changedSep 29, 2026

removedOutput schema / properties / last_activity / properties / last_remote

Removed value: -{
-  "description": "Last call relayed from a web client (source cloud), when there was one.",
-  "type": "object"
-}

changedOutput schema / properties / tcc / description

Previous value: -"TCC permission states. Automation keys (<app>_automation) are one of notDetermined | authorized | denied | notRunning."New value: +"TCC permission states, e.g. granted | denied | authorized."

removedOutput schema / properties / tcc / properties / mail_automation_last_known

Removed value: -{
-  "description": "Present only when mail_automation is notRunning: \"authorized\" if the grant was confirmed before. Not a live verdict.",
-  "type": "string"
-}

removedOutput schema / properties / tcc / properties / messages_automation
```
Removed value: -{
-  "type": "string"
-}
```
removedOutput schema / properties / tcc / properties / notes_automation
```
Removed value: -{
-  "type": "string"
-}
```

removedOutput schema / properties / tcc / properties / omnifocus_automation

Removed value: -{
-  "description": "Each <app>_automation is present only if the app is installed; <app>_last_known and <app>_note may accompany it.",
-  "type": "string"
-}

removedOutput schema / properties / web_agents

Removed value: -{
-  "description": "Web (cloud-relay) clients, counts only: `count` connected (absent when unknown), `relayed_calls_since_launch`, `last_remote_call` {at, tool, ok, ago_s}.",
-  "type": "object"
-}

Changed7 schema fields changedSep 26, 2026

addedOutput schema / properties / last_activity / properties / last_remote

Added value: +{
+  "description": "Last call relayed from a web client (source cloud), when there was one.",
+  "type": "object"
+}

changedOutput schema / properties / tcc / description

Previous value: -"TCC permission states, e.g. granted | denied | authorized."New value: +"TCC permission states. Automation keys (<app>_automation) are one of notDetermined | authorized | denied | notRunning."

addedOutput schema / properties / tcc / properties / mail_automation_last_known

Added value: +{
+  "description": "Present only when mail_automation is notRunning: \"authorized\" if the grant was confirmed before. Not a live verdict.",
+  "type": "string"
+}

addedOutput schema / properties / tcc / properties / messages_automation
```
Added value: +{
+  "type": "string"
+}
```
addedOutput schema / properties / tcc / properties / notes_automation
```
Added value: +{
+  "type": "string"
+}
```

addedOutput schema / properties / tcc / properties / omnifocus_automation

Added value: +{
+  "description": "Each <app>_automation is present only if the app is installed; <app>_last_known and <app>_note may accompany it.",
+  "type": "string"
+}

addedOutput schema / properties / web_agents

Added value: +{
+  "description": "Web (cloud-relay) clients, counts only: `count` connected (absent when unknown), `relayed_calls_since_launch`, `last_remote_call` {at, tool, ok, ago_s}.",
+  "type": "object"
+}

Changed1 schema field changedAug 7, 2026

addedOutput schema / properties / skill_capture

Added value: +{
+  "description": "Dark-launch counters for repeated-workflow detection (user-generated skills, PR-1). would_fire_signatures = how many times a save-this-workflow nudge WOULD have fired; ring_size = entries in the recent-calls ring. PRIVACY: the ring is IN-MEMORY ONLY — never written to disk, never transmitted, cleared on restart — and argument values are PII-scrubbed on entry, so it holds the SHAPE of a workflow, not its content. No UI acts on it yet; it exists to tune thresholds. Opt out with the skill_capture_enabled config flag.",
+  "properties": {
+    "ring_size": {
+      "type": "integer"
+    },
+    "would_fire_signatures": {
+      "type": "integer"
+    }
+  },
+  "type": "object"
+}

Changed2 schema fields changedJul 30, 2026

addedOutput schema / properties / build_sha

Added value: +{
+  "description": "Git short SHA of the build (F-041). 'unknown' if unstamped.",
+  "type": "string"
+}

addedOutput schema / properties / built_at

Added value: +{
+  "description": "UTC build timestamp (F-041). 'unknown' if unstamped.",
+  "type": "string"
+}

Changed1 schema field changedJul 28, 2026

changedOutput schema / properties / services / description

Previous value: -"Per-domain reachability summary (mail, calendar, contacts, teams, onedrive, slack, …); shape varies by domain. May include `scan_pending: true` right after startup, meaning the per-service running/accounts values are placeholders and not yet authoritative."New value: +"Per-domain reachability summary (mail, calendar, contacts, teams, onedrive, slack, …); shape varies by domain. May include `scan_pending: true` right after startup, meaning the per-service running/accounts values are placeholders and not yet authoritative. Once scanned, `scanned_seconds_ago` gives the age (seconds) of that background snapshot and `freshness` restates that a false/0/not-available is advisory, not a live check — never gate a task on it."

Changed1 schema field changedJul 27, 2026

changedOutput schema / properties / services / description

Previous value: -"Per-domain reachability summary (mail, calendar, contacts, teams, onedrive, slack, …); shape varies by domain."New value: +"Per-domain reachability summary (mail, calendar, contacts, teams, onedrive, slack, …); shape varies by domain. May include `scan_pending: true` right after startup, meaning the per-service running/accounts values are placeholders and not yet authoritative."

First observedJun 30, 2026

TDQS

A4.5/5.0

Behavior5/5

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

Annotations already declare readOnly/openWorld false, but the description goes well past them: sub-500ms latency, passive behaviour (never prompts, never opens windows, never touches network), and two state-machine caveats — scan_pending placeholders vs. scanned_seconds_ago snapshot cadence (~60s) — plus the staleness warning. This is exactly the context an agent needs to interpret a 'false' service value without panicking.

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?

Front-loaded with the payload contents, then the caveats, which is the right order. Slightly apologetic/repetitive at the end — 'never use this tool as a preflight gate' and 'the actual tool call is the source of truth, just attempt it' say the same thing twice — but no sentence is wasted.

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?

Output schema exists, so return values need no exposition; annotations cover the safety profile; the description supplies the remaining operational knowledge (freshness, placeholder semantics, no-side-effects guarantee) that an agent would otherwise get wrong. Complete for a zero-param read tool that many agents will call defensively.

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?

Zero parameters, so there is nothing for the description to disambiguate and the baseline of 4 applies. It does add semantic meaning to key output fields (scan_pending, scanned_seconds_ago), which is bonus value even though the output schema technically covers it.

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

Purpose4/5

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

Specific verb ('Returns a structured snapshot') plus an enumerated inventory of what the snapshot contains (versions, AI client, relay state, TCC permissions, per-service reachability). It clearly separates itself from report_problem, but the near-neighbour diagnostic siblings (lmcp_doctor, permissions_status, run_diagnostics, list_missing_permissions) are never named, so an agent must infer the boundary from their names alone.

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-to-use ('verify the environment is healthy before attempting a tool, or to understand what's installed and accessible'), explicit when-NOT ('never use this tool as a preflight gate to skip or cancel a task'), and an explicit alternative for a competing need ('For reporting failures, use report_problem instead'). Nothing about 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 Local MCP 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.

LMCP State

Input Schema

Output Schema

Schema Changelog

TDQS

Glama MCP Gateway

Resources