asc-aso-audit
Pass
Audited by Gen Agent Trust Hub on Jun 14, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands using the
ascCLI tool to fetch app tags and information from App Store Connect. - [EXTERNAL_DOWNLOADS]: The skill integrates with the Astro MCP, which communicates with external APIs to provide keyword rankings, competitor analysis, and suggestions.
- [PROMPT_INJECTION]: The skill processes untrusted metadata content from local files, creating a surface for indirect prompt injection.
- Ingestion points: Reads subtitle, name, keywords, and description from local JSON files in the
./metadatadirectory. - Boundary markers: No specific delimiters or safety instructions are defined to wrap the untrusted metadata during processing.
- Capability inventory: Performs shell command execution through the
ascCLI and makes API-based tool calls via Astro MCP. - Sanitization: There are no instructions for sanitizing or escaping the content of the metadata fields before they are used in analysis or reporting.
Audit Metadata