integrate-agentforce-android

Pass

Audited by Gen Agent Trust Hub on Oct 26, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill configures the project to fetch Maven dependencies from well-known services and Salesforce-owned repositories, including JitPack and Amazon S3 buckets.
  • [COMMAND_EXECUTION]: Performs automated modifications to project configuration files, including build.gradle.kts and settings.gradle.kts, and scaffolds multiple Kotlin source files for infrastructure implementation.
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests user-provided values like Salesforce domains and Agent IDs to populate code templates.
  • Ingestion points: User responses during interactive discovery phases defined in SKILL.md.
  • Boundary markers: Uses {{PLACEHOLDERS}} within the Kotlin snippets in the references/snippets/ directory to delimit user data.
  • Capability inventory: File creation and modification tools are used across the Android project structure.
  • Sanitization: The instructions guide the agent to use // TODO: comments for missing values and ensure that placeholders are substituted correctly before final file generation.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 26, 2026, 01:50 PM