The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to run swift build and swift test. These commands compile and execute code found in the local repository, which could lead to code execution if the repository contains malicious Swift code.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to analyze and modify Swift source code, creating an attack surface for indirect prompt injection. 1. Ingestion points: Reads Swift source files and package manifests from the local repository. 2. Boundary markers: There are no explicit instructions or delimiters to prevent the agent from following instructions embedded within the processed code. 3. Capability inventory: The skill uses swift build and swift test which are capable of executing arbitrary code during the compilation or testing phases. 4. Sanitization: The skill does not implement any sanitization or validation of the code before processing it.

swift-concurrency