The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes the Bash and Task tools to perform tech stack detection and execute parallel security auditing agents. These capabilities are consistent with the skill's primary purpose of conducting a thorough security review of a repository.
[PROMPT_INJECTION]: The skill presents an indirect prompt injection surface as it ingests untrusted content from the codebase being audited.
Ingestion points: The Read, Glob, and Grep tools are used to ingest files across the entire codebase, including package.json, configuration files, and source code.
Boundary markers: The skill does not explicitly define boundary markers or instruct the sub-agents to ignore instructions found within the audited data.
Capability inventory: The skill has access to powerful tools including Bash, Write, and Task, which could be targeted by a malicious codebase.
Sanitization: There is no evidence of sanitization or escaping of the ingested code before it is passed to the parallel security agents.

security-audit