mcp-loop
Pass
Audited by Gen Agent Trust Hub on Aug 14, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands such as
flutter runandflutter pub getthrough the Flutter MCP server. - Evidence: The 'Vòng lặp' section specifies running
flutter runand hot reloads, and the 'Phân loại lỗi' table mentions executingflutter pub get. - [INDIRECT_PROMPT_INJECTION]: The skill establishes an automated loop that processes untrusted data from build logs and runtime output to perform code modifications, creating a surface for indirect prompt injection.
- Ingestion points: Build logs, runtime logs, and widget tree output ingested in steps 2, 3, and 5 of the 'Vòng lặp' section.
- Boundary markers: The skill does not define specific delimiters or instructions to ignore embedded commands within the logs.
- Capability inventory: The agent has the ability to modify source files and execute shell commands based on the log analysis.
- Sanitization: No sanitization or validation of the log content is described before it is used to drive the fix loop.
Audit Metadata