The Agent Skills Directory

[Indirect Prompt Injection] (LOW): The skill utilizes mcp__context7__query-docs to fetch external documentation, creating an indirect prompt injection surface. External documentation content is untrusted and could attempt to manipulate the agent's behavior. Ingestion points: External documentation data from mcp__context7__query-docs. Boundary markers: No delimiters or ignore-embedded-instruction warnings are present in the skill instructions. Capability inventory: The skill has access to powerful tools including Bash, Edit, and Write. Sanitization: No evidence of validation or sanitization of documentation content prior to processing.
[Data Exposure & Exfiltration] (SAFE): Code examples for sensitive values like JWT signing keys correctly use variables (e.g., jwtSigningKey) instead of hardcoded strings or secrets.
[Command Execution] (SAFE): While the Bash tool is permitted, the skill does not contain any suspicious shell commands or scripts. All provided examples are standard .NET C# code snippets.

aspire