$ loading_
Query Perfetto traces in natural language for focused performance analysis.
The available materials describe an open-source local MCP server for Perfetto trace analysis, with no required secrets and no declared remote endpoints, and no obvious high-risk red flags. The main considerations are its local execution capability as an MCP tool and its likely access to user-provided trace data, so it should be used in a least-privilege environment.
The materials explicitly state that no keys or environment variables are required, and no API tokens, account credentials, or external service authorizations are mentioned, so credential exposure and abuse risk appears low.
No remote host endpoints are declared, and the description does not indicate that traces or prompts are sent to external services; based on the available materials, there is no clear data egress path.
The objective checks indicate it has executes-code capability; as an MCP server, it will typically run local processes and handle analysis requests. This is a normal capability for this type of tool, and the current materials do not show any abnormal system privileges beyond its stated purpose.
Its stated function is to get answers from Perfetto traces, so it likely needs to read user-provided trace files or related analysis data. The materials do not specify write locations, and there is no clear sign of overbroad access, but it should be assumed to handle potentially sensitive performance or behavioral data.
The source is an open-source GitHub repository under Apache-2.0 with about 192 stars, which provides some auditability and community adoption—both positive signals. However, the lack of a README and unknown maintenance status reduce visibility into implementation details and dependency hygiene, so caution is appropriate rather than a high-risk rating.
Copy the install command and let the AI configure it · recommended for beginners
No copy-paste install info for "perfetto-mcp" yet — see the docs or source repo.
Analyze this Perfetto trace and identify the biggest jank points during app startup, summarizing the causes by thread, function, and time range.
A summary of key startup bottlenecks, related threads and call segments, and a readable explanation of the causes.
Inspect this Perfetto trace, explain which processes and threads consumed the most CPU, and point out the time ranges with abnormal spikes.
A ranked list of high CPU consumers, abnormal time windows, and the corresponding analysis findings.
Using this Perfetto trace, find the intervals with the worst frame drops and explain whether scheduling, I/O, or main-thread blocking caused them.
The dropped-frame intervals, likely root cause categories, and a concise explanation for engineering investigation.
Analyze Linux perf data and pinpoint performance bottlenecks with typed commands.
Let AI assistants query Perplexity for cited web answers and filtered search.
Monitor dev logs and surface runtime errors for instant AI code verification.
Inspect database schemas, index issues, table bloat, and query plans.
Collect web performance metrics and generate normalized reports for issue diagnosis.
Let AI query and analyze OpenTelemetry traces to debug apps faster.