The Agent Skills Directory

[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because its workflow requires agents to read and follow the 'true state' of a task from a Joplin note to resume work. Content within these notes could potentially contain malicious instructions that the agent might follow.
Ingestion points: Untrusted data enters the agent context via the mcp__joplin__read_note and mcp__joplin__search_notes tools as described in SKILL.md.
Boundary markers: The task note structure defined in TASK-FORMAT.md does not include delimiters or instructions to ignore embedded commands within the note body.
Capability inventory: The agent has the ability to manage Joplin notebooks and notes, and can execute system commands like date as mentioned in SKILL.md.
Sanitization: There is no evidence of sanitization or filtering of the note content in SKILL.md or TASK-FORMAT.md before the agent processes it.

task-notes