The Agent Skills Directory

[PROMPT_INJECTION]: The skill demonstrates an indirect prompt injection surface by retrieving untrusted text data (tweets and user descriptions) from the X API.
Ingestion points: get_user_tweets.py and search_tweets.py fetch external content into the agent context.
Boundary markers: No delimiters or isolation warnings are used to separate retrieved content from system instructions.
Capability inventory: The skill possesses write capabilities (posting tweets and uploading media) which could be targeted by instructions found in processed tweets.
Sanitization: No sanitization or safety filtering is applied to the retrieved text content before it is presented to the agent.
[COMMAND_EXECUTION]: The skill functionality is implemented through local Python scripts. post_tweet.py performs a dynamic import of the upload_media.py script from its local directory to facilitate file uploads. This is a legitimate internal orchestration pattern and does not involve untrusted remote code.

x-api