crm-archaeology

Pass

Audited by Gen Agent Trust Hub on Aug 30, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes several local command-line tools to perform its functions, including pac (Power Platform CLI) for solution management, dotnet for tool installation, az (Azure CLI) for authentication token retrieval, and rg (ripgrep) for static analysis of minified JavaScript bundles.
  • [EXTERNAL_DOWNLOADS]: The skill guides the user to install the Microsoft.PowerApps.CLI.Tool via dotnet tool install. This is an official, well-known utility provided by Microsoft for interacting with Power Platform environments.
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and document metadata from external Dynamics 365 environments, which represents a potential attack surface.
  • Ingestion points: Environment metadata, solution XML files (e.g., Customizations.xml), and JavaScript web resources are retrieved via Web API and solution unpacking as described in references/extraction-queries.md.
  • Boundary markers: The skill does not implement explicit delimiters or safety instructions when processing external metadata for documentation assembly.
  • Capability inventory: The skill possesses the capability to execute shell commands (pac, rg), perform file system writes (solution unpacking to workspace), and conduct network operations (Web API GET requests).
  • Sanitization: The skill focuses on faithful extraction and documentation of the environment and does not mention specific sanitization routines for the ingested metadata.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 30, 2026, 02:26 AM
Security Audit — agent-trust-hub — crm-archaeology