process-api-net11
Pass
Audited by Gen Agent Trust Hub on Nov 6, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill documents and promotes the use of System.Diagnostics.Process APIs for executing external commands and capturing output, which increases the agent's capability for command execution.
- Ingestion points: No direct ingestion points are defined in the static content of the skill, but the documentation teaches the agent to use these APIs for future tasks that may involve project data or user input.
- Boundary markers: The documentation and code examples do not provide instructions on using boundary markers or delimiters to separate untrusted data from command arguments.
- Capability inventory: The skill enables extensive process orchestration capabilities, including starting processes, capturing stdout/stderr, and handle inheritance via the described .NET 11 APIs in SKILL.md.
- Sanitization: There is no mention of argument sanitization or validation in the provided documentation, which could lead to command injection if the agent adopts these code patterns for processing external strings.
Audit Metadata