adp-workforce

Warn

Audited by Gen Agent Trust Hub on Apr 29, 2026

Risk Level: MEDIUMEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONREMOTE_CODE_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill instructs users to install the Membrane CLI globally via npm install -g @membranehq/cli@latest. This fetches the tool from the public npm registry.- [COMMAND_EXECUTION]: The skill provides instructions for the agent to execute various shell commands using the membrane CLI, including authentication, connection management, and action execution.- [REMOTE_CODE_EXECUTION]: The command membrane action create enables the creation of new functionality based on natural language descriptions. This involves the dynamic generation and deployment of code within the Membrane platform environment.- [DATA_EXFILTRATION]: The skill accesses and processes highly sensitive HR and payroll data (e.g., demographics, pay distributions) from ADP Workforce Now via the Membrane API integration.- [PROMPT_INJECTION]: The skill features a surface for indirect prompt injection by processing external data from API responses.
  • Ingestion points: External data enters via membrane action run and membrane action list outputs.
  • Boundary markers: No explicit delimitation or instructions to ignore embedded commands are present in the documentation.
  • Capability inventory: The agent has the ability to execute shell commands and interact with the CLI as specified in SKILL.md.
  • Sanitization: No sanitization or validation logic is defined for the external data ingested from API actions.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Apr 29, 2026, 08:42 PM