The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill requires the agent to process untrusted data from the codebase and error logs, which can influence the agent's logic and subsequent command execution. \n
Ingestion points: Data ingestion occurs through read_file and search_files calls on the local project directory. \n
Boundary markers: The instructions do not specify the use of delimiters or ignore-rules for handling content from external files. \n
Capability inventory: The agent has access to the terminal tool for shell execution and file modification capabilities. \n
Sanitization: No explicit steps are provided for sanitizing or validating input data before it affects agent behavior. \n- [COMMAND_EXECUTION]: The skill utilizes the terminal tool to execute standard development commands, such as pytest for testing and git for version history analysis. These operations are intended for local debugging and align with the skill's purpose.

systematic-debugging