The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses the infsh CLI via bash to interact with the Twitter/X API. All shell execution is restricted to the infsh tool by the platform's allowed-tools configuration, ensuring the agent operates within the intended scope.
[EXTERNAL_DOWNLOADS]: Directs users to installation guides and documentation hosted on the official GitHub repository and the inference.sh domain. These are standard vendor resources for the platform's ecosystem.
[PROMPT_INJECTION]: The skill has a surface for indirect prompt injection as it facilitates posting content to social media that may originate from untrusted external sources.
Ingestion points: Processes user-provided JSON input for tweet text and media URLs, as seen in the examples in SKILL.md.
Boundary markers: Absent. The instructions do not define delimiters or provide warnings to the agent regarding the handling of potentially malicious embedded instructions in the content to be posted.
Capability inventory: Provides high-impact capabilities including posting tweets, sending direct messages, following users, and deleting content via the infsh CLI.
Sanitization: No sanitization or validation logic is described to handle untrusted input before it is passed to the automation apps.

twitter-automation