The Agent Skills Directory

[SAFE]: The skill implements a standard debugging workflow including error capture, reproduction, and isolation. It uses authorized tools for file manipulation and shell execution.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze untrusted runtime data such as error messages, logs, and stack traces (SKILL.md). While this provides a theoretical surface for indirect prompt injection, it is the primary and intended function of the skill. The instructions maintain a focus on technical diagnosis and minimal fixes, which mitigates the risk of the agent following malicious instructions embedded in logs.
[EXTERNAL_DOWNLOADS]: The skill mentions utilizing 'exa search' and 'context7 mcp' for research. These references guide the agent to use available tool-calling interfaces for information gathering and do not involve unauthorized external downloads or remote code execution from unknown sources.

debugger