The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted text provided by users as its primary function. It lacks defined boundary markers or instructions to treat the input text as data only, which could allow instructions within the source text to influence the agent. The skill has the capability to write to external note-taking tools, specifically mentioning the 'Scratchpad connector'. There is no mention of sanitizing or validating the input text to prevent malicious instructions from being executed during the summarization or saving process.

summary-writer