The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes system commands such as grep, awk, javac, java, and python3 to search the codebase, compile Java solutions, run test drivers, and execute local validation scripts.
[DYNAMIC_EXECUTION]: To verify solution correctness, the skill dynamically generates a temporary Java driver file in the /tmp/ directory, which is then compiled and executed locally.
[INDIRECT_PROMPT_INJECTION]: The skill has an attack surface for indirect prompt injection as it processes untrusted data from external sources.
Ingestion points: Reads code drafts from leetcode_java/src/main/java/ws/Workspace26.java or directly from user-provided input snippets.
Boundary markers: The skill does not define specific delimiters or instructions to ignore embedded prompts within the drafts.
Capability inventory: The skill possesses the ability to write/edit files, execute shell commands, and compile/run Java code.
Sanitization: There is no explicit sanitization or filtering applied to the ingested draft content, though the skill mandates a rigid output structure which acts as a partial constraint.
[EXTERNAL_DOWNLOADS]: The skill references documentation and problem metadata from well-known services (LeetCode and NeetCode) and the author's official GitHub repository (yennanliu/CS_basics).

lc-java