Flutter Agent Lens

Unclaimed

by dhruvanbhalara

Agent-first Model Context Protocol (MCP) server to debug, profile, and optimize running Flutter apps. Connects AI assistants to Dart VM Service over WebSockets for direct access to widget trees, memory snapshots, network traffic, stateful CPU profiling, and layout diffing.

Unclaimed listing

Is this your MCP server?

This listing was auto-indexed from the public record. Claim it to edit the page, set compatibility and unlock growth tools. Takes under two minutes.

Claim this server

Tools (18)

connection

connect<br>connectdtd<br>disconnect

diagnose_project

bundlesize<br>deeplinks

discover_apps

N/A

duration_seconds

in consolelogs (watch), network (watch), memory (getmemorytimeline, watchgcpressure): Sets sampling or monitoring duration in seconds (default: 5.0, max: 30).

exclude_flutter_widgets

in rebuildtracking (action: stop / getcounts): Excludes built-in Flutter/SDK and dependency widgets from the rebuild list (default: true).

filter

in consolelogs (action: watch): Optional text filter substring for live logs.

forceGC

in memory (action: getsnapshot / save / diffallocations): Runs garbage collection before inspection (default: true/false).

include_details

in network (watch / stop / getprofile): Includes full request & response headers, cookies, and bodies (default: false).

includeExtensions

in getappinfo: Hides the full service extensions list when false (default: false).

includeRawNode

in widget (action: inspect): Hides raw JSON layout nodes when false (default: false).

includeRawResponse

in memory and network: Hides raw JSON payloads when false (default: false).

limit

in diagnoseproject (action: deeplinks): Sets max deep links checked.

maxDepth

in widget (action: gettree): Sets widget tree depth (default: 8).

platform

in diagnoseproject (action: deeplinks): Platform target (e.g. android or ios, required).

projectOnly

in widget (action: gettree): Filters out non-user-project widgets (default: true).

set_response_format

N/A

slow_threshold_ms

in network (action: watch): Threshold in ms to flag slow requests (default: 500).

topN

in memory (action: compare): Sets max class differences returned (default: 10).