new-api-support
Pass
Audited by Gen Agent Trust Hub on Feb 23, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection via the <entity_name> argument.
- Ingestion points: The <entity_name> input enters via the skill's argument hint and is used in SKILL.md.
- Boundary markers: None present; the argument is not wrapped in delimiters or safety instructions.
- Capability inventory: The skill executes shell commands (find, grep) and Swift tests (swift test, xcodebuild).
- Sanitization: None present; the input is used directly in string interpolation for shell execution.
- [COMMAND_EXECUTION]: User-provided input is interpolated into shell commands, creating a risk of command injection.
- Evidence: Step 1 uses
grep -r "<entity_name>"and Step 4 usesswift test --filter "testInvestigate_<EntityName>". A malicious user could provide a crafted string to execute arbitrary code on the developer's machine.
Audit Metadata