The Agent Skills Directory

[PROMPT_INJECTION]: The skill facilitates the ingestion of live, untrusted data from X (formerly Twitter), including tweets, profiles, and community posts. This represents a surface for indirect prompt injection where malicious instructions embedded in social media content could be interpreted by the agent as commands.
Ingestion points: Data is fetched via x_read, x_search, x_timeline, x_lists, and x_communities endpoints from twitr.sh.
Boundary markers: The instructions do not specify any delimiters or safety markers to isolate external data from the system prompt.
Capability inventory: The agent has the ability to perform network requests and execute financial transactions via the mcp__agentcash__fetch tool.
Sanitization: No evidence of content filtering or escaping for the retrieved social media data is present.
[EXTERNAL_DOWNLOADS]: The skill communicates with an external, non-whitelisted domain twitr.sh to retrieve data and process 402 Payment Required status codes.
[COMMAND_EXECUTION]: The skill utilizes an external tool dependency, AgentCash MCP, to perform network operations and handle USDC payments. This grants the agent the capability to interact with a financial payment gateway autonomously, though instructions advise user confirmation for costs exceeding $0.10.

x-twitter-data