The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill facilitates the installation of the x-twitter-pp-cli and x-twitter-pp-mcp binaries. It recommends using npx with the @mvanhorn/printing-press package and go install from github.com/mvanhorn/printing-press-library. These resources are provided by the skill's author.
[COMMAND_EXECUTION]: The skill's functionality is built around executing shell commands, specifically the x-twitter-pp-cli binary. It also provides a Python script to extract GraphQL hashes from HAR files and update local configuration.
[DATA_EXFILTRATION]: The CLI tool includes a --deliver webhook:<url> feature that allows the agent to send command results, which may contain user data or API responses, to a specified URL. Additionally, the feedback command supports sending local data to an external endpoint via the X_FEEDBACK_ENDPOINT environment variable.
[PROMPT_INJECTION]: The skill is subject to indirect prompt injection risks.
Ingestion points: The agent retrieves untrusted data from the X platform, such as tweets, direct messages, and chat content (e.g., via tweets search-posts-recent or chat get-conversation).
Boundary markers: There are no instructions for using delimiters or protective prompts to prevent the agent from following instructions hidden within the data retrieved from X.
Capability inventory: The CLI can perform file writes (--deliver file:<path>) and network POST requests (--deliver webhook:<url>), which could be abused if the agent is manipulated by injected content.
Sanitization: The skill does not describe any mechanisms for sanitizing or validating the data retrieved from external APIs before processing.

pp-x-twitter