analyze-canvas-performance
Pass
Audited by Gen Agent Trust Hub on Jun 20, 2026
Risk Level: SAFE
Full Analysis
- [Local Source Analysis]: The skill analyzes local
.pa.yamlfiles to identify performance bottlenecks, such as N+1 database calls and unoptimized formulas. This analysis is conducted locally and is necessary for the skill's primary function as an auditing tool. - [Indirect Prompt Injection Surface]: As the skill processes user-provided application code and formulas, it possesses a surface for indirect prompt injection. Maliciously crafted comments or formulas within the analyzed app could attempt to influence the agent's report; however, the skill's use of a predefined technical reference guide and structured audit rules helps mitigate this concern.
- [Local File Generation]: The skill generates a local HTML report to present its findings. This report includes helpful links to official documentation for remediation. The file is stored on the local filesystem and does not involve network exfiltration.
- [MCP Tool Integration]: The skill utilizes dedicated platform tools such as
sync_canvasandget_appchecker_errors. This is a standard and safe practice for integrating with the Power Apps ecosystem to ensure data is synchronized and validated correctly. - [Automated Remediation]: Upon user request, the skill can apply fixes to the YAML files. This process includes a validation step via the
compile_canvastool to ensure that the modifications do not introduce syntax or logic errors into the application.
Audit Metadata