The Agent Skills Directory

[PROMPT_INJECTION]: Indirect Prompt Injection Surface (Category 8). The skill retrieves untrusted data from social media and news sources, which can contain embedded instructions designed to manipulate the AI agent.
Ingestion points: Data is ingested from the CT Monitor API via endpoints such as /tweets/feed, /info/feed, and /twitter/realtime as documented in SKILL.md.
Boundary markers: The skill instructions do not specify the use of delimiters or provide directives to ignore embedded instructions within the ingested content.
Capability inventory: The skill employs curl for networking and jq for data processing as defined in SKILL.md.
Sanitization: There are no logic checks, escaping mechanisms, or validation steps defined to sanitize the content of tweets or news articles before they are processed by the LLM.
[EXTERNAL_DOWNLOADS]: Fetches market data, briefings, and social media signals from the external domain https://api.ctmon.xyz/api.
[COMMAND_EXECUTION]: Utilizes system-level binaries curl and jq to interact with remote services and process JSON results.

ct-monitor