The Agent Skills Directory

[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because its primary function is to ingest and process untrusted external code and documentation for auditing.
Ingestion points: Uses Read, Grep, Glob, and Bash to pull external content into the agent's context.
Boundary markers: The instructions do not define clear delimiters or instructions to ignore embedded commands within the data being audited.
Capability inventory: The agent has access to Bash, Write, Edit, and TeamCreate, providing a significant impact surface if a malicious instruction in the audited code is followed.
Sanitization: No sanitization or validation of the input data is specified before it is processed by the agent.
[COMMAND_EXECUTION]: The skill explicitly allows the use of the Bash tool. While this is likely intended for running legitimate audit utilities or static analysis tools, the ability to execute arbitrary shell commands poses a risk of unintended system modification or unauthorized data access if not strictly monitored.

agent-audit