The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses several shell commands to perform its functions, specifically jstar init, jstar review, jstar chat, and git add. These are part of the core functionality provided by the vendor's toolset.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it analyzes external code which may contain untrusted instructions.
Ingestion points: Local repository files are ingested when the agent runs git add . followed by jstar review or jstar review --json.
Boundary markers: There are no explicit delimiters or warnings instructed to the agent to ignore instructions embedded within the code being reviewed.
Capability inventory: The agent has the capability to execute shell commands (jstar, git) and read/write files as part of the fix cycle.
Sanitization: No explicit sanitization or validation of the code content is mentioned before it is processed by the review logic.

code-review