The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes standard shell utilities including find, jq, sed, and grep to perform read-only structural analysis of project files. These operations are restricted to the local Assets directory.
[DATA_EXFILTRATION]: No network activity or access to sensitive system locations (such as ~/.ssh, ~/.aws, or .env files) was detected.
[PROMPT_INJECTION]: The instructions are clear and instructional, without any attempts to bypass agent safety protocols, system prompts, or ethical constraints.
[INDIRECT_PROMPT_INJECTION]: The skill has an ingestion surface for untrusted data from project .asmdef files.
Ingestion points: Assets/*.asmdef (processed in SKILL.md via find and jq).
Boundary markers: Absent; the skill does not use specific delimiters to wrap external content.
Capability inventory: Limited to read-only shell commands and echo for reporting results.
Sanitization: Uses jq for structured field extraction, providing implicit validation of the JSON format.

asmdef-lint