mdcp-ux

Pass

Audited by Gen Agent Trust Hub on Aug 4, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructs the agent to discover and execute validation commands from the repository's developer documentation or package scripts. While this is a standard workflow for software development skills, it creates a dependency on the integrity of the project's local configuration files.
  • [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection attack surface (Category 8) by ingesting untrusted data from user-provided work items and external documentation paths.
  • Ingestion points: WORK_ITEM (user input) and the location specified in WORK_ITEM_LOOKUP (external or local file paths).
  • Boundary markers: The skill does not explicitly define delimiters or instructions to ignore embedded commands within the ingested data.
  • Capability inventory: The agent is authorized to perform Git operations (branching, committing) and execute shell commands discovered in the project environment.
  • Sanitization: No explicit sanitization or validation of the external content is mentioned before it is used to influence the agent's task planning and execution logic.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 4, 2026, 09:49 PM
Security Audit — agent-trust-hub — mdcp-ux