Agent Skills by DragonKid
Showing 17 vetted skills indexed across 1 GitHub repositories.
dragonkidtodo
Automate TODO management via Telegram /todo commands with a local Markdown file.
dragonkidgit-config-separation
Separate personal git settings from shared configs using Git's include mechanism.
dragonkidguide
Load and summarize plugin, workflow, and database MCP reference files.
dragonkidresearch-staged
Guide multi-stage research workflows with approval gates between stages.
dragonkidprompt-improver
Rewrite ambiguous prompts into clear, specific instructions for AI models.
dragonkidprompt-generator
Converts task descriptions or existing prompts into structured system prompts for LLMs.
dragonkidfastapi-lifespan-di
Initialize and tear down async resources in FastAPI via lifespan context managers.
dragonkidpydantic-multi-source-settings
Consolidate Python app configuration from env, .env, and remote services.
dragonkiddto-factory-method
Convert inconsistent API responses into typed Pydantic domain models.
dragonkidchoosing-task-vs-main-session
Choose between Task-based subagents and main-session workflows for complex operations.
dragonkidllm-agent-token-audit
Audit LLM agent prompts and workspace injections to quantify and prune token waste.
dragonkidllm-parse-retry
Parse LLM outputs into valid JSON with retry logic and fallback.
dragonkidpydantic-basesettings-class-attributes
Prevent non-annotated attribute errors in Pydantic BaseSettings subclasses.
dragonkidllm-agent-tool-calling-optimization
Enforce parallel tool call batches with explicit numeric budgets for LLM agents.
dragonkidpython-project-toolchain
Automate Python project toolchain setup with uv, ruff, pyright, pytest, and pre-commit hooks.
dragonkidorganizing-dotfiles-configs
Organize dotfiles configurations and plan symlink strategies with a decision framework.
dragonkidoptimize-llm-config-docs
Optimize verbose LLM configuration documents into compact actionable guidance.