The Agent Skills Directory

[SAFE]: The skill consists entirely of high-level instructions for a debugging methodology. It does not automate any actions, perform network requests, or access sensitive files.
[INDIRECT_PROMPT_INJECTION]: The protocol involves analyzing external data such as stack traces and logs. This constitutes an attack surface for indirect prompt injection if the analyzed data contains malicious instructions. However, the skill does not grant the agent unsafe capabilities to act on such data, and this risk is inherent to the debugging use case.
[COMMAND_EXECUTION]: The text mentions tools like curl and git log as examples for the agent to suggest to the user during debugging. There are no automated command executions or scripts included in the skill.

debug-protocol