The Agent Skills Directory

[COMMAND_EXECUTION]: The skill's frontmatter contains shell hooks (pre and post) that execute commands upon skill activation and completion. These hooks navigate to /workspaces/ruvector/.claude/intelligence and execute a Node.js CLI script (cli.js) to process files.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from external sources, including agent communication logs, task results, and metrics, to generate performance reports and optimization recommendations.
Ingestion points: Data is ingested via npx claude-flow bottleneck detect and the mcp__claude-flow__task_results tool.
Boundary markers: No specific boundary markers or instructions to ignore embedded commands within the processed data were identified.
Capability inventory: The skill can execute shell commands via hooks and write reports to the file system using the --output flag.
Sanitization: There is no evidence of sanitization or escaping of the ingested agent logs before they are interpolated into the generated reports.
[DYNAMIC_EXECUTION]: The documentation provides examples of custom JavaScript analysis scripts that utilize child_process.exec to dynamically execute CLI commands and process their output at runtime.
[EXTERNAL_DOWNLOADS]: The skill frequently references the use of npx claude-flow, which triggers the download and execution of the claude-flow package from the NPM registry if it is not already present in the environment.

performance-analysis