compare-test-runs
Pass
Audited by Gen Agent Trust Hub on Sep 16, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to utilize various subcommands of the
tuistCLI tool to retrieve test run data, module lists, and suite breakdowns. These commands (e.g.,tuist test show,tuist test list) are standard operations for the tool and are used here to gather comparison data in JSON format. - [INDIRECT_PROMPT_INJECTION]: The skill processes external data in the form of test results and metadata (names, failure messages, and logs) provided by the
tuistCLI. This creates a surface where malicious content embedded in test output could theoretically attempt to influence the agent's analysis or summary. - Ingestion points: Output from
tuist test case run show,tuist test case run list, and related commands as identified inSKILL.md. - Boundary markers: No specific delimiters or warnings for the agent to ignore embedded instructions in the test output are present.
- Capability inventory: The skill is limited to executing
tuistCLI commands for data retrieval; it does not perform file writes or network operations directly. - Sanitization: The skill does not define specific sanitization steps for the tool output, relying on the agent's default safety guardrails for processing external data.
Audit Metadata