agkan
Pass
Audited by Gen Agent Trust Hub on Oct 18, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted task data, including titles, bodies, and comments, which provides a surface for indirect prompt injection. Maliciously crafted task content could attempt to override the agent's behavior during planning or execution phases.
- Ingestion points: Task bodies (which can be read from arbitrary local files via the
--fileflag), task comments, metadata keys/values, and JSON files processed during import operations. - Boundary markers: There are no explicit instructions or delimiters defined to warn the agent to ignore instructions embedded within the task content it retrieves.
- Capability inventory: The skill provides capabilities for file system interaction (export/import, reading task bodies from files), process management (starting/stopping a local board server), and task lifecycle management.
- Sanitization: The skill does not describe any sanitization or validation logic to filter out potential prompt injection strings from task-related data before it is presented to the agent.
Audit Metadata