The Agent Skills Directory

[DYNAMIC_EXECUTION]: The skill uses node -e to execute complex JavaScript logic for managing the OAuth loopback redirect, including setting up a temporary local HTTP server to capture the authentication token.
[COMMAND_EXECUTION]: Shell commands are used to read local configuration files, retrieve git remote information, and launch the system's default browser for the authentication process.
[INDIRECT_PROMPT_INJECTION]: The skill presents an attack surface for indirect prompt injection by processing untrusted data that influences subsequent code execution.
Ingestion points: The skill reads from project-specific configuration files (.claude-plugin/project-config.json), user-level configuration (~/.aem/ops-config.json), and raw user input when parsing project URLs.
Boundary markers: No delimiters or safety instructions are used when processing the ingested data before it is passed to other tools.
Capability inventory: The skill has access to shell command execution (bash), dynamic JavaScript evaluation (node -e), and filesystem write access.
Sanitization: The use of basic string splitting (cut, awk) to extract organization and site names is insufficient to prevent the inclusion of JavaScript metacharacters that could modify the logic of the dynamically generated node -e script.

auth