The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses local shell commands via bun run to start performance capture processes, such as bun run react-scan:capture and bun run react-scan:capture:brand. These are standard vendor-specific development scripts.
[INDIRECT_PROMPT_INJECTION]: The skill processes external data that could theoretically be manipulated to influence agent behavior.
Ingestion points: The agent is instructed to call get_react_scan_recording and analyze the resulting summary and raw events from the React application (SKILL.md).
Boundary markers: No explicit delimiters or instructions are provided to the agent to treat the profiling data as untrusted or to ignore embedded instructions.
Capability inventory: The skill environment allows for shell command execution and interaction with browser tools (SKILL.md).
Sanitization: The instructions do not specify any validation or sanitization steps for the data returned by the profiling tools before the agent performs its analysis.

react-scan