hig-project-context
Pass
Audited by Gen Agent Trust Hub on Jun 28, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection because it reads untrusted data from various project files and uses it to populate a configuration file that guides subsequent agent behavior. \n
- Ingestion points: Content is read from README.md, Package.swift, .xcodeproj, Info.plist, and source code files as instructed in SKILL.md. \n
- Boundary markers: The instructions in SKILL.md lack explicit delimiters or instructions for the agent to ignore potentially malicious embedded prompts within the ingested files. \n
- Capability inventory: The agent is empowered to read multiple file types and write to the .claude/apple-design-context.md file as described in SKILL.md. \n
- Sanitization: No procedures for escaping or validating the content extracted from the project files are defined in SKILL.md. \n- [COMMAND_EXECUTION]: The skill involves the execution of the grep command to search the local filesystem for accessibility-related code patterns, as specified in SKILL.md.
Audit Metadata