The Agent Skills Directory

[REMOTE_CODE_EXECUTION]: The skill provides instructions to download and install binary components via npx -y @mvanhorn/printing-press-library and go install github.com/mvanhorn/printing-press-library/.... These resources are provided by the skill author.
[DATA_EXFILTRATION]: The CLI tool supports a --deliver webhook:<url> parameter that transmits command results to an external web address. This functional feature provides a mechanism for data exfiltration if the agent is directed to use a malicious or unauthorized URL.
[INDIRECT_PROMPT_INJECTION]: The skill fetches content from the Braze REST API, including marketing campaigns, templates, and content blocks. This content represents an untrusted data source that could contain malicious instructions for the agent.
Ingestion points: Data retrieved via various braze-pp-cli commands like campaigns list or templates get-email-info.
Boundary markers: No explicit delimiters or instructions to ignore embedded commands are provided in the skill prompts.
Capability inventory: The skill uses the Read Bash tool and executes the braze-pp-cli binary which can access network resources.
Sanitization: The instructions do not specify any validation or sanitization of the data retrieved from the API.
[COMMAND_EXECUTION]: The skill utilizes the Read Bash tool to execute braze-pp-cli commands. While necessary for the skill's function, it grants the agent the ability to run shell commands on the underlying system.

pp-braze