The Agent Skills Directory

[SAFE]: The skill operates as intended for research and information gathering, using search tools to retrieve content from specified social and technical platforms and formatting it into structured markdown.
[PROMPT_INJECTION]: Indirect Prompt Injection Surface. The skill processes content from third-party websites which may contain instructions intended to manipulate the agent's summaries. 1. Ingestion points: External content fetched via web search from various social platforms and blogs (SKILL.md). 2. Boundary markers: Absent; no specific instructions are provided to the agent to distinguish between data and potentially malicious embedded instructions in the search results. 3. Capability inventory: Web search and text summarization. 4. Sanitization: Absent; the skill relies on the underlying model's safety and filtering mechanisms.

topic-collector