ti-howtos
Pass
Audited by Gen Agent Trust Hub on Mar 8, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [SAFE]: The skill consists of markdown-based documentation and technical references for mobile development with the Titanium SDK, covering topics such as location services, maps, push notifications, and media APIs.- [SAFE]: References to external tools such as Fastlane, Appium, and various Node.js packages are contextual to standard development workflows and do not involve the skill executing unauthorized remote code.- [PROMPT_INJECTION]: The skill creates a surface for indirect prompt injection by directing the agent to ingest and analyze untrusted project-specific files.\n
- Ingestion points: Documentation in 'SKILL.md' specifies that 'tiapp.xml' and folders like 'app/' and 'Resources/' serve as indicators for project detection and guidance generation.\n
- Boundary markers: Absent; the instructions do not define delimiters or warnings for the agent to ignore instructions embedded within the analyzed files.\n
- Capability inventory: The skill provides access to 'Read', 'Grep', 'Glob', 'Edit', 'Write', 'Bash', and 'Bash(node *)', which grant the agent significant capabilities if influenced by a poisoned input file.\n
- Sanitization: None; the skill lacks logic to escape or validate the contents of the 'tiapp.xml' or other project files before they are processed in the agent context.
Audit Metadata