The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructions define a 'Memory Protocol' that requires the agent to run the cat command on a local file (.claude/context/memory/learnings.md) to maintain persistent state between sessions.
[PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface due to its interaction with external code. Ingestion points: The skill is intended to ingest and review Java source code provided by users. Boundary markers: There are no delimiters or specific instructions to prevent the agent from being influenced by malicious commands embedded within the reviewed code. Capability inventory: The agent is configured with access to powerful tools, specifically Bash, Write, Edit, Grep, and Glob. Sanitization: The skill lacks validation or sanitization routines to ensure that processed code does not contain malicious instructions for the agent.

java-expert