The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from Twitter (tweets, user profiles, bios, etc.) which could contain malicious instructions designed to influence the agent's behavior.
Ingestion points: Tools such as get_twitter_user_tweets, search_twitter, search_twitter_advanced, and get_twitter_deleted_tweets fetch content from external sources (SKILL.md).
Boundary markers: The skill does not explicitly define or enforce delimiters or instruction-ignore warnings for the fetched content.
Capability inventory: The skill possesses network communication capabilities to ai.6551.io (api_client.py).
Sanitization: No evidence of input sanitization or validation of the fetched Twitter content is provided in the documentation.
[EXTERNAL_DOWNLOADS]: The installation process requires cloning an external repository and using the uv package manager to sync and run the code, which involves downloading and executing third-party code locally.
[COMMAND_EXECUTION]: The skill configuration and execution rely on running shell commands (e.g., uv run, claude mcp add) which allows for arbitrary command execution within the context of the server's setup and operation.

opentwitter-mcp-server