tailrocks-swift-project-remediate
Pass
Audited by Gen Agent Trust Hub on Sep 8, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from local project files and tool outputs while maintaining the capability to execute shell commands.
- Ingestion points: Reads existing native macOS project files and project-specific templates referenced in SKILL.md.
- Boundary markers: Includes specific instructions in references/runtime-trust.md to ignore embedded instructions in repository files and registry content.
- Capability inventory: Executes subprocesses for toolchains including xcodebuild, swiftlint, xcodegen, and mise tasks as detailed in references/lint-and-format.md and references/testing.md.
- Sanitization: Implements a policy of bounding commands, network, and output, and explicitly prohibits the reading or copying of discovered credentials.
- [COMMAND_EXECUTION]: The skill requires the agent to execute a variety of development tools to perform its primary function of project remediation.
- Evidence: Instructions include running xcrun swift-format, swiftlint, periphery scan, xcodegen, and xcodebuild with specific flags and configurations across multiple reference files.
- [DYNAMIC_EXECUTION]: The skill involves template-based generation and execution of scripts.
- Evidence: Mentions copying bytes from canonical templates to generate project files and scripts like window-id.swift and capture.sh, then executing them as part of the build and verification process.
Audit Metadata