wayfinder
Pass
Audited by Gen Agent Trust Hub on Oct 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [SAFE]: The skill utilizes a 'disable-model-invocation' configuration, ensuring it is only triggered manually by the user, which prevents unauthorized autonomous execution.
- [INDIRECT_PROMPT_INJECTION]: The agent ingests external data from an issue tracker. This potential attack surface is addressed by strict process rules requiring human interaction (HITL) for key decisions and prohibiting the agent from answering its own questions.
- [COMMAND_EXECUTION]: The skill executes a local script 'frontier.py' to query the state of the task ledger. This is a functional requirement for the skill and operates within the local environment.
Audit Metadata