The Agent Skills Directory

[SAFE]: The skill is designed for debugging and root-cause analysis. It does not contain any prompt injection, obfuscation, or data exfiltration patterns. The instructions focus on hypothesis-driven diagnosis rather than arbitrary code execution.
[COMMAND_EXECUTION]: The skill utilizes standard, safe commands such as git bisect, grep, and node --version for environmental and version verification. These operations are restricted to the local repository and are appropriate for the skill's stated purpose.
[INDIRECT_PROMPT_INJECTION]: The skill possesses an inherent attack surface as it is designed to process external, potentially untrusted data like error logs, stack traces, and repository content. This is a functional requirement for a debugger and is categorized as low risk.
Ingestion points: Reads error logs, stack traces, screenshots, and repository source code.
Boundary markers: Not explicitly defined in the instructions for data processing.
Capability inventory: Includes grep -rn across the repository and git bisect for regression testing.
Sanitization: Relies on the user and the agent's logic to interpret logs rather than automated execution of log-embedded code.

hunt