Rustam
Community@rustams · Moscow
Автоматизирую бизнес-процессы. Немного увлекаюсь SPA, WPA.
Agent Skills by Rustam
Showing 25 vetted skills indexed across 1 GitHub repositories.
develop-b24-php
Develop Bitrix24 backend applications with PHP, Symfony, and the Bitrix24 PHP SDK.
multi-agent-patterns
Design multi-agent systems with supervisor, swarm, and hierarchical patterns.
bdi-mental-states
Transform RDF knowledge graphs into BDI mental states for agent reasoning.
context-degradation
Diagnose context degradation patterns like lost-in-middle and context poisoning in LLMs.
implement-b24-features
Implement Bitrix24 UI widgets, automation robots, event handlers, and RabbitMQ task queues.
develop-b24-python
Develop Bitrix24 backend applications with Python, Django, and b24pysdk.
context-compression
Compress AI conversation histories to reduce token usage and prevent context limits.
bitrix24-static-local-app
Build and package Bitrix24 static local app archives from Nuxt.js projects.
memory-systems
Compare production memory frameworks and design persistence architectures for agent memory systems.
product-manager-skills
Generate PRDs, user stories, roadmaps, and positioning statements for product management.
advanced-evaluation
Implement LLM-as-a-Judge evaluation pipelines with bias mitigation and rubric design.
navigate-b24-project
Navigates the Bitrix24 Starter Kit monorepo to locate frontend, backend, and infrastructure code files.
develop-b24-frontend
Develop Bitrix24 frontend applications with Nuxt 3 and Pinia.
manage-b24-environment
Manage Bitrix24 Docker services, Cloudpub tunnels, and backend databases.
context-fundamentals
Explain context engineering concepts for AI agent systems.
develop-b24-node
Develop Bitrix24 backend applications with Node.js, Express, and the Bitrix24 JS SDK.
Bitrix24 MCP server
Discover Bitrix24 REST API methods, parameters, and examples via MCP server.
project-development
Guide LLM project initiation from task evaluation to pipeline architecture and cost estimation.
inbound-agent-system
Manage AI agents for the Inbound Marketing Bitrix24 project with memory and context rules.
evaluation
Evaluate AI agent performance using multi-dimensional rubrics and LLM-as-judge.
frontend-design
Generate production-grade frontend interfaces with bold aesthetic direction.
context-optimization
Optimizes transformer context windows by applying selective compression and caching techniques to manage token usage efficiently.
filesystem-context
Manage and retrieve filesystem context using ls, glob, grep, and read_file.
tool-design
Design and organize AI agent tools with interface and architecture guidance.