dataverse-web-api
Warn
Audited by Gen Agent Trust Hub on Oct 16, 2026
Risk Level: MEDIUMPRIVILEGE_ESCALATIONDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [PRIVILEGE_ESCALATION]: The skill's instructions in
SKILL.mdandresources/best-practices.mdexplicitly direct the agent to run PowerShell scripts with the-ExecutionPolicy Bypassflag. This action circumvents local security controls intended to prevent the execution of unsigned or potentially malicious scripts.\n- [DYNAMIC_EXECUTION]: The skill operates by generating and executing local PowerShell scripts at runtime to interact with the Dataverse API. The reliance on dynamically created executable content increases the potential impact of script injection or local resource abuse.\n- [INDIRECT_PROMPT_INJECTION]: The skill creates a vulnerability surface for indirect prompt injection during schema generation.\n - Ingestion points: Ingests user-supplied metadata definitions (tables, columns, relationships) which the agent then uses to build API payloads.\n
- Boundary markers: No delimiters or safety instructions are used to separate the user-provided data from the agent's executable instructions.\n
- Capability inventory: The agent has the capability to perform highly privileged schema and data modifications on a remote Dataverse environment.\n
- Sanitization: The skill lacks logic for escaping or validating user inputs before they are included in the generated PowerShell code or OData requests.\n- [COMMAND_EXECUTION]: The skill utilizes command-line tools like the Azure CLI (
az account get-access-token) to manage authentication. While targeting trusted Microsoft services, the execution of shell commands to handle sensitive credentials represents a security-sensitive operations pattern.
Audit Metadata