The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests and analyzes external project source code, which serves as a potential vector for indirect prompt injection if the code contains malicious instructions designed to influence the agent.
Ingestion points: The skill reads package.json and performs recursive searches across files in the src/ directory (SKILL.md).
Boundary markers: Absent. There are no specific delimiters or instructions provided to the agent to distinguish between its own logic and instructions that might be embedded in the analyzed code.
Capability inventory: The skill has the ability to execute shell commands (grep, wc) and write files to the local filesystem (.frontend-review/report/latest/md/performance-review.md).
Sanitization: Absent. No explicit sanitization or filtering of the ingested source code content is performed before analysis.
[COMMAND_EXECUTION]: The skill uses shell commands to identify performance patterns and usage metrics in the codebase.
Evidence: Employs grep and wc to find instances of React memoization hooks, virtualization components, and concurrent features in *.tsx and *.ts files.

frontend-review-performance