ticket
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes
mkdir -p tmp/tickets/doneto initialize its local storage directory. This is a standard and safe file system operation. - [INDIRECT_PROMPT_INJECTION]: The skill acts as a data bridge, taking user input and writing it into Markdown files in
tmp/tickets/which are then processed by an autonomous tool named 'Ralph'. - Ingestion points: User-provided task descriptions, reference numbers, and acceptance criteria collected during the interactive session in
SKILL.md. - Boundary markers: The skill uses basic Markdown headers as structure but provides no explicit delimiters or instructions to the downstream agent to ignore potentially malicious embedded commands.
- Capability inventory: The skill itself performs file system writes. The downstream consumer ('Ralph') likely has broader capabilities such as command execution or code modification to fulfill the tickets.
- Sanitization: No sanitization, escaping, or validation of the user-provided content is performed before it is written to the ticket files.
Audit Metadata