The Agent Skills Directory

[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests and processes data from external, potentially untrusted web pages.
Ingestion points: Uses browser-devtools-cli to navigate to external URLs and retrieve performance data, HTTP requests, and console messages (SKILL.md).
Boundary markers: No specific boundary markers or instructions to ignore embedded commands within the analyzed content are provided.
Capability inventory: Features capabilities for browser navigation, console message retrieval, screenshot capture, and Node.js process debugging via node-devtools-cli.
Sanitization: There is no evidence of sanitization or filtering of the content retrieved from external sites before it is processed by the agent.
[COMMAND_EXECUTION]: The skill uses specialized CLI tools to perform system-level performance monitoring and debugging.
Evidence: Employs node-devtools-cli to attach to Node.js processes using a Process ID (--pid) and to set tracepoints at specific source code locations (SKILL.md). These capabilities are characteristic of debugging tools and are consistent with the skill's primary objective of backend performance auditing.

performance-audit