The Agent Skills Directory

Shell Command Interpolation: The skill generates and executes shell commands that incorporate user-provided variables, such as search terms (term), connector identifiers (api-id), and dataset/table names. This pattern represents a potential surface for command injection if inputs are not properly sanitized before being passed to the bash environment.
Dynamic Execution Patterns: The workflow includes the execution of a Node.js one-liner to validate configuration files and triggers npm run build. These operations execute logic based on the local project environment and configuration, which is standard for build tools but introduces a dependency on the integrity of the workspace files.
Indirect Prompt Injection Surface: The agent processes output from the ms connector list CLI tool to provide recommendations to the user. This creates a pathway for external data to influence agent behavior.
Ingestion points: Output from CLI commands ms connector list and ms connector list-actions (Step 2a).
Boundary markers: The skill does not define specific delimiters or instructions to ignore instructions embedded in the CLI output.
Capability inventory: The skill has access to Bash, Write, Edit, and AskUserQuestion tools.
Sanitization: There is no explicit sanitization or filtering logic applied to the data retrieved from the CLI tool before it is presented or acted upon.

add-data-source