trello
Warn
Audited by Gen Agent Trust Hub on Feb 16, 2026
Risk Level: MEDIUMPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [Prompt Injection] (MEDIUM): The skill is susceptible to indirect prompt injection (Category 8) because it fetches and processes data from an external source (Trello).
- Ingestion points: Untrusted data enters the agent's context through
curlcommands that fetch card descriptions (desc) and comments (text) inSKILL.md. - Boundary markers: Absent. The skill does not provide instructions to the agent to delimit or ignore instructions embedded within the Trello data.
- Capability inventory: The skill possesses network-write capabilities via
curlto create cards, move cards, and post comments. - Sanitization: Absent. The data is piped through
jqbut the resulting strings are processed directly by the agent without escaping or validation. - [Command Execution] (LOW): The skill relies on
curlfor all operations. While these target the legitimate Trello API, the use of template placeholders like{boardId}and{listId}requires the agent to ensure these inputs are properly sanitized to prevent local command injection if an attacker can control those ID values.
Audit Metadata