The Agent Skills Directory

[PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface through the ingestion of external content.\n
Ingestion points: The skill reads data from the Unity console via the read_console tool and examines project state via unity_reflect in SKILL.md.\n
Boundary markers: There are no specific instructions or delimiters provided to the agent to treat console log content as untrusted data or to ignore potential instructions embedded within those logs.\n
Capability inventory: The agent has significant capabilities to modify the environment, including create_script, script_apply_edits, and apply_text_edits, and can execute code via manage_editor and execute_code as detailed in SKILL.md and references/profiling-guide.md.\n
Sanitization: No sanitization or filtering of the ingested console logs or reflection data is performed before the agent processes it.\n- [COMMAND_EXECUTION]: The skill utilizes tools to execute code directly within the Unity Editor environment.\n
Execution method: The execute_code tool is described for performing specialized tasks such as verifying texture compression settings via TextureImporter in references/profiling-guide.md.

unity-debug-profiler