The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill defines a workflow where the agent ingests and analyzes external data including application source code, performance traces (JSON), Lighthouse reports, and database query logs. This represents an attack surface for indirect prompt injection if these external sources are maliciously crafted. However, this is an inherent risk of the agent's primary function and no specific execution vulnerabilities are introduced.
Ingestion points: Source code files, Lighthouse trace files, and database query log output.
Boundary markers: None explicitly defined in the skill instructions.
Capability inventory: The agent retains file system modification and command execution capabilities (via npx).
Sanitization: No explicit sanitization of ingested performance data or logs is specified.
[COMMAND_EXECUTION]: The skill recommends the use of standard development tools for performance monitoring and enforcement, such as lhci (Lighthouse CI) and bundlesize. These are executed via npx in a CI/CD context and are used for their intended purpose within the skill's scope.

performance-optimization