The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes npx eslint to perform automated code audits. This is a standard industry practice for identifying performance and code quality issues.
[DYNAMIC_EXECUTION]: The skill instructions direct the agent to generate temporary ESLint configuration files (e.g., eslint.config.mjs) at runtime to apply specialized performance rules. This is a common pattern for focused linting tasks.
[EXTERNAL_DOWNLOADS]: The skill recommends installing specific Node.js packages (e.g., eslint-plugin-react-perf, eslint-plugin-react-native) if they are missing from the project environment to support the linting phase.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted project source code and external data sources (network logs, profiler reports) which creates a potential surface for indirect prompt injection.
Ingestion points: The agent reads project files (.js, .ts, .tsx) in references/lint-rules.md and references/semantic-checklist.md, and network logs in references/fix-reference.md.
Boundary markers: None identified in the provided instructions to differentiate between instructions and data.
Capability inventory: The skill has capabilities for shell command execution (npx), file system operations (writing config files), and device interaction (argent-device-interact).
Sanitization: No evidence of input sanitization or escaping of external content before processing.

argent-react-native-optimization