The Agent Skills Directory

[PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface as it ingests and processes untrusted data from the local environment.\n
Ingestion points: The skill reads local Swift test files and captures output from xcodebuild and git status to display results.\n
Boundary markers: No explicit delimiters or instructions are used to separate the ingested data from the agent's operational instructions.\n
Capability inventory: The skill has the capability to execute system commands via Bash, read files from the filesystem, and write reports to the .agents/research/ directory.\n
Sanitization: There is no evidence of sanitization, escaping, or validation of the content read from files or captured from command outputs.\n- [COMMAND_EXECUTION]: The skill performs several command-line operations using the Bash tool to automate the development workflow.\n
Evidence: It executes xcodebuild for running tests, git for repository status and commits, and xcrun simctl or killall for simulator management.\n
Safety features: Destructive actions like file deletion (clearing DerivedData) or committing changes are protected by the AskUserQuestion tool, requiring explicit user consent before execution.\n
Context: All commands are standard for iOS/macOS development and target specific, expected directories and services.

run-tests