The Agent Skills Directory

[Indirect Prompt Injection] (HIGH): The skill is designed to analyze enterprise workflow patterns and requests high-privilege tools (Bash, Write, Edit). This configuration is highly susceptible to indirect prompt injection, where data processed by the agent contains hidden instructions that the agent may obey with system-level privileges.
Ingestion points: Enterprise workflow data, patterns, and best practices analyzed via the Read tool (implied by description and allowed-tools).
Boundary markers: None identified in the SKILL.md file; no delimiters or 'ignore embedded instructions' warnings are present.
Capability inventory: Bash, Write, Edit, Read (declared in SKILL.md).
Sanitization: No sanitization, validation, or filtering logic is defined for the external content to be processed.
[Command Execution] (HIGH): The skill explicitly requests the Bash tool. In an enterprise context involving code generation and validation, this provides a direct path for executing arbitrary shell commands if the agent is manipulated by adversarial input.
[No Code] (INFO): No executable scripts or implementation logic (e.g., .py, .js, .sh) were provided for analysis. This assessment is based on the declarations and metadata found in SKILL.md.

impact-analysis-helper