testany-case
Pass
Audited by Gen Agent Trust Hub on Oct 15, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill requires the agent to execute shell commands, specifically
curl, provided by MCP tools to fetch execution logs and credential metadata. This architectural pattern for data retrieval involves the agent executing dynamically constructed command strings from tool outputs.\n- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from the Testany platform, such as test scripts within ZIP packages and execution logs. These inputs lack sanitization or protective boundary markers, presenting a risk of indirect prompt injection.\n - Ingestion points: Content of test scripts (via
testany_get_case_script) and execution logs (viatestany_get_dry_run_log).\n - Boundary markers: Absent; there are no instructions to use delimiters or treat retrieved content as data rather than instructions.\n
- Capability inventory: The agent can execute shell commands (
curl) and manage platform resources through tools liketestany_create_case,testany_update_case, andtestany_delete_case.\n - Sanitization: No validation or filtering is applied to scripts or logs before they are processed by the agent.
Audit Metadata