The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from ClickUp task names, descriptions, and comments and has write capabilities that could be exploited by malicious content in those fields.
Ingestion points: Data enters the agent context through clickupfy task get, clickupfy task search, and clickupfy comment list in SKILL.md.
Boundary markers: There are no explicit delimiters or instructions for the agent to ignore potentially malicious embedded content in the ingested data.
Capability inventory: The agent can update tasks, delete tasks, and manage time entries, providing a significant surface for an injection to act upon.
Sanitization: The instructions suggest verifying changes after the fact but do not include proactive sanitization or filtering of external input strings.

clickupfy-dev