The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The instructions recommend the use of npx @claude-flow/cli@latest, which fetches and executes code from the npm registry.
[INDIRECT_PROMPT_INJECTION]: The skill implements a workflow for learning from task completions, which introduces a potential surface for indirect prompt injection from processed task data.
Ingestion points: Task context and steps are recorded using mcp__plugin_ruflo-core_ruflo__hooks_intelligence_trajectory-start and mcp__plugin_ruflo-core_ruflo__hooks_intelligence_trajectory-step.
Boundary markers: There are no instructions provided to the agent regarding delimiters or safety warnings for the ingested task data.
Capability inventory: The skill uses Bash and specialized tools for training, optimizing, and compressing neural patterns.
Sanitization: No data sanitization or validation steps are mentioned before the trajectory data is processed for learning.
[COMMAND_EXECUTION]: The skill uses the Bash tool to execute MCP tool commands and external CLI utilities.

neural-train