The Agent Skills Directory

[REMOTE_CODE_EXECUTION]: The skill invokes npx @claude-flow/cli@latest to download and execute code from the NPM registry at runtime during the execution of pre-task and post-task hooks.
[COMMAND_EXECUTION]: User-provided arguments (<context-name> and <aggregate-name>) are interpolated directly into shell command strings used with npx. This lacks shell escaping or sanitization, creating a command injection vulnerability where malicious input could execute arbitrary code.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted input from $ARGUMENTS and passes it to high-capability tools like Bash and file writing tools without adequate boundary markers or sanitization.
Ingestion points: User-provided <context-name> and <aggregate-name> arguments.
Boundary markers: None present; the input is used directly in prompts and commands.
Capability inventory: The skill uses Bash for command execution and Write/Edit for file system modifications.
Sanitization: None detected; arguments are used raw in shell command strings and file paths.
[EXTERNAL_DOWNLOADS]: The skill initiates downloads of the @claude-flow/cli package from an external source at runtime without integrity checks or source verification.

ddd-aggregate