The Agent Skills Directory

[PROMPT_INJECTION]: The skill uses high-pressure language ("A hefty fee will be applied if you violate them") in references/xcode-project-setup.md to enforce specific operational constraints. This is a behavioral override technique designed to influence the agent's reasoning by introducing artificial consequences.
[INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data that is used in sensitive operations, creating an attack surface for indirect prompt injection.
Ingestion points: User-supplied dataSourceID and PRODUCT_BUNDLE_IDENTIFIER read from Xcode project files (SKILL.md).
Boundary markers: Absent. Values are interpolated directly into shell command strings for curl operations.
Capability inventory: The skill performs shell command execution (curl) and project modification via a Swift script (main.swift).
Sanitization: Absent. The instructions do not specify validation or shell-escaping for the IDs, which could lead to command injection if malicious strings are provided as input.
[EXTERNAL_DOWNLOADS]: The automated setup tool (scripts/xcode_spm_setup/Package.swift) fetches the XcodeProj library from the tuist organization on GitHub. This is a well-known and standard dependency in the iOS development community for manipulating project files.
[COMMAND_EXECUTION]: The skill requires the execution of shell commands (curl) to perform its mandatory project healthcheck and uses the swift run command to execute its internal project configuration tool.

contentsquare-ios-sdk