The Agent Skills Directory

[COMMAND_EXECUTION]: The skill provides instructions for the agent to execute shell commands using npx to perform performance analysis tasks. Key commands include npx claude-flow bottleneck detect for issue identification and npx claude-flow analysis performance-report for metrics summaries.
[EXTERNAL_DOWNLOADS]: The skill uses npx, which fetches the claude-flow package from the npm registry for execution. This is a standard development practice for using Node.js-based utilities.
[PROMPT_INJECTION]: The skill has an indirect prompt injection surface due to its analysis of external task results and swarm communication logs.
Ingestion points: Data is processed through metrics analysis commands and MCP tool results like mcp__claude-flow__task_results (SKILL.md).
Boundary markers: No explicit delimiters or boundary markers are defined to separate ingested metrics from agent instructions.
Capability inventory: The skill can execute shell commands (npx), perform file writes (--export, --output), and apply system optimizations (--fix).
Sanitization: There is no documentation of explicit sanitization for the ingested performance metrics.

performance-analysis