
setup
Diagnose the current mode and CLI availability when using upbit-strategy-toolkit for the first time or when another skill cannot run. Checks uv/uvx behavior in clone mode and agent skills mode. Environment setup, getting started, first use, setup, uv explanation. Use when the user is starting for the first time, or when another skill detects unmet environment requirements (missing uv, uv sync failure, CLI execution failure, etc.) and recommends routing here.
- Installs
- 118
- Last 24h
- —
- Trending rank
- —
- Security
- Not audited
Install
Works with every agent the skills CLI supports.
npx skills add https://github.com/upbit-official/upbit-strategy-toolkit --skill setup- 01
Install
Run the command in your project. The CLI asks which agents to install for.
- 02
Describe the task
Ask your agent for something this skill covers. It loads setup automatically when relevant.
- 03
Review the result
The agent follows the SKILL.md steps, scripts and references bundled with the skill.
SKILL.md
The instructions your agent reads when the skill is triggered.
setup
Identify the upbit-strategy-toolkit usage mode and diagnose whether the CLI can run in the current environment. This skill is a mode-detection and diagnostic utility, not installation automation. create-strategy and backtest handle strategy creation and backtesting.
OS-specific installation instructions for cases where uv is missing are split into references/:
- uv installation reference: references/uv-install.md
Triggers
- "Use it for the first time"
- "How do I set up the environment?"
- "What is uv? / It says uv is missing"
- Another skill (create-strategy/backtest) detected unmet environment requirements (missing uv,
uv syncfailure, CLI execution failure, etc.) and recommended routing here
Automatic routing recommendation: if create-strategy / backtest fails to run the wrapper on its first call, recommend setup.
Routing Contract
- Role: environment diagnosis and recovery entry point
- Next step:
create-strategy - Recovery entry: guide to this skill when
create-strategyorbacktestdetects missing uv,uv syncfailure, or CLI execution failure - Out of scope: this skill does not handle strategy design, backtest interpretation, or market-rule work
Do Not
- Do not change the user's system Python arbitrarily - uv isolates the required version in a virtual environment. Do not guide the user to upgrade/downgrade system Python.
- Do not recommend
pip install -g- clone-mode dependencies stay inside the virtual environment throughuv sync. - Do not connect to live trading - see Live Trading Guardrail.
Workflow
Check pass/fail for each item before moving to the next. On failure, report as factual statements which command failed and what error occurred.
setup
[ ] 1. Identify usage mode
[ ] 2. Check uv installation
[ ] 3. Run mode-specific CLI diagnostics
1. Identify Usage Mode
Check whether the current workspace is the upbit-strategy-toolkit repo.
| Mode | Detection | Default data root (used unless config set data-dir was run) |
|---|---|---|
| clone mode | Current cwd or an ancestor has name = "upbit-strategy-toolkit" in pyproject.toml and upbit_strategy_toolkit/ | The current working directory at run time |
| agent skills mode (global) | Skill was installed with npx skills add ... -g; the wrapper script's own path sits directly under $HOME/.claude/skills, $HOME/.agents/skills, or $HOME/.cursor/skills | The current working directory at run time |
| agent skills mode (project-scoped) | Skill was installed with npx skills add ... (no -g); the wrapper script's own path does not sit under the home-directory skills paths above, and the current workspace does not contain the upbit-strategy-toolkit repo | The current working directory at run time |
When neither UPBIT_TOOLKIT_DATA_DIR nor a stored config set data-dir value is present, the CLI resolves the data root to the current working directory at run time, regardless of mode. If the user has run config set data-dir, that stored value is used instead. Report the resolved value with config get data-dir rather than assuming it from the mode table above.
2. Check uv Installation
uv --version
If it prints normally, this passes. If uv is missing, point to references/uv-install.md, then check uv --version again after installation.
3-A. Clone Mode Diagnostics
Run from the repo workspace/cwd.
uv sync
uv run upbit-strategy-toolkit --help
uv sync creates .venv/ and prepares the required Python and dependencies. Clone mode diagnostics pass when uv run upbit-strategy-toolkit --help prints help.
On failure, report factual details: which package failed and what error occurred. Do not bypass with arbitrary pip install.
3-B. Agent Skills Mode (Global or Project) Diagnostics
When the repo is not cloned, check by running the CLI directly from the GitHub source. This step is the same for global and project-scoped installs.
uvx --from git+https://github.com/upbit-official/upbit-strategy-toolkit.git upbit-strategy-toolkit --help
Agent skills mode diagnostics pass when help is printed. On failure, report based on the original error whether it is missing uv, GitHub access failure, or Python 3.12 acquisition failure.
Bidirectional Handoff
After Setup Completion -> create-strategy
When diagnostics pass, naturally hand off to the next skill with this line:
upbit-strategy-toolkit CLI diagnostics complete. You can now create a strategy.
Next step: create-strategy skill
- Start with natural language such as "Create a golden cross strategy".
- The generated {data_root}/strategies/{slug}.json can be validated with the backtest skill.
Blocked During Setup -> Factual Report
Without judging good/bad, state exactly which step produced which error:
[Step 2 blocked] `uv --version` returned "command not found"
-> Shell restart or PATH check is required (see references/uv-install.md)
[clone mode Step 3 blocked] pandas wheel build failed during `uv sync`
Original error: ...
-> Recheck Python 3.12+ availability. If system Python is 3.11 or lower, run `uv python install 3.12` and retry.
[agent skills mode Step 3 blocked] GitHub access failed during `uvx --from ... upbit-strategy-toolkit --help`
Original error: ...
-> Network or GitHub access permissions need to be checked.
Live Trading Guardrail
Same policy as create-strategy / backtest. Reject live-trading integration requests even during setup.
Response template: If the user's message is in Korean, use the Korean version; otherwise use the English version.
Korean:
이 도구는 백테스트(과거 데이터 시뮬레이션) 전용입니다.
실전 매매가 필요하다면 [upbit-agent-skills](https://github.com/upbit-official/upbit-agent-skills)를 이용해주세요.
setup 스킬은 백테스트 환경 설정만 지원합니다.
English:
This tool is for backtesting (historical data simulation) only.
Use the separate [upbit-agent-skills](https://github.com/upbit-official/upbit-agent-skills) tool for live trading.
The setup skill only supports backtest environment setup.
Execution Rule Summary
- Identify the mode first, then run only that mode's diagnostic commands
- clone mode:
uv --version,uv sync,uv run upbit-strategy-toolkit --help - agent skills mode (global or project-scoped):
uv --version,uvx --from git+https://github.com/upbit-official/upbit-strategy-toolkit.git upbit-strategy-toolkit --help - Do not guide changes to system Python - uv isolates it in a virtual environment
- Report failures only as factual statements (step + command + original error)
- Naturally hand off to create-strategy after setup completes
- Refuse live trading requests and point to upbit-agent-skills
FAQ
What does the setup skill do?
Diagnose the current mode and CLI availability when using upbit-strategy-toolkit for the first time or when another skill cannot run. Checks uv/uvx behavior in clone mode and agent skills mode. Environment setup, getting started, first use, setup, uv explanation. Use when the user is starting for the first time, or when another skill detects unmet environment requirements (missing uv, uv sync failure, CLI execution failure, etc.) and recommends routing here.
How do I install setup?
Run npx skills add https://github.com/upbit-official/upbit-strategy-toolkit --skill setup in your project. The skills CLI detects the agents you use and copies the skill into the right folder. For Claude Code you can also place the skill folder at ~/.claude/skills/setup/ (personal) or .claude/skills/setup/ (project).
Which AI agents can use setup?
Any agent that supports the open Agent Skills format (a folder with a SKILL.md file), including Claude Code, Codex, Cursor, GitHub Copilot, Gemini CLI, Windsurf, Cline, OpenCode and more.
Is setup safe to install?
No partner security audit is available yet. Skills can run commands and scripts, so read the SKILL.md and any bundled files before installing.
Who maintains setup?
It is published by upbit-official in upbit-official/upbit-strategy-toolkit. It has 118 recorded installs.
More from upbit-official/upbit-strategy-toolkit
backtest
upbit-official/upbit-strategy-toolkit
Open the skill to read its SKILL.md instructions.
create-strategy
upbit-official/upbit-strategy-toolkit
Open the skill to read its SKILL.md instructions.
Popular ai & agents skills
google-agents-cli-publish
google/agents-cli
This skill should be used when the user wants to "publish an agent", "publish my ADK agent", "register an agent with Gemini", "publish to Gemini" (formerly Gemini Enterprise), or needs guidance on the agents-cli publish gemini command. Also use when the user wants to "manage agents in Agent Registry", "list/update/delete registered agents", or "register an MCP server". Covers ADK vs A2A registration modes, programmatic and interactive usage, flag reference, auto-detection from deployment metadata, Agent Registry fleet management, and troubleshooting. Part of the agents-cli skills suite. Do NOT use for deployment (use google-agents-cli-deploy).
google-agents-cli-eval
google/agents-cli
This skill should be used when the user wants to "run an evaluation", "evaluate my agent", "evaluate my ADK agent", "write an eval dataset", "analyze eval failures", "compare eval results", "optimize agent", or needs guidance on the Gemini platform eval methodology and the Quality Flywheel. Covers eval metrics, dataset schema, LLM-as-judge scoring, and common failure causes. Applies to any agents-cli project, whatever framework the agent is written in. Do NOT use for agent API code patterns (ADK: use google-agents-cli-adk-code), deployment (use google-agents-cli-deploy), or project scaffolding (use google-agents-cli-scaffold).
skill-creator
anthropics/skills
Create new skills, modify and improve existing skills, and measure skill performance. Use when users want to create a skill from scratch, edit, or optimize an existing skill, run evals to test a skill, benchmark skill performance with variance analysis, or optimize a skill's description for better triggering accuracy.
codex-pet
prime-skills/runcomfy-agent-skills
Codex Pet generator on RunComfy. Build a Codex-compatible Codex Pet spritesheet.webp + pet.json from a single reference image, drop it into ${CODEXHOME:-$HOME/.codex}/pets/ / and Codex picks it up as a custom Codex Pet next to the 8 built-ins. This skill produces the exact Codex Pet atlas Codex expects (1536x1872 PNG/WebP, 8 cols x 9 rows, 192x208 cells, 9 animation states — idle, running-right, running-left, waving, jumping, failed, waiting, running, review). Calls OpenAI GPT Image 2 edit ONCE via the local RunComfy CLI as runcomfy run openai/gpt-image-2/edit to produce a canonical Codex Pet pose, then assembles all 9 animation rows programmatically with ImageMagick micro-transforms — no Codex Pro, no $imagegen, no OPENAIAPIKEY required, only RUNCOMFYTOKEN. Triggers on "codex pet", "create codex pet", "make codex pet", "hatch codex pet", "/hatch image", "desktop pet codex", "codex pets", "spritesheet.webp", or any explicit ask to build a custom pet for OpenAI Codex.
full-output-enforcement
leonxlnx/taste-skill
Overrides default LLM truncation behavior. Enforces complete code generation, bans placeholder patterns, and handles token-limit splits cleanly. Apply to any task requiring exhaustive, unabridged output.
executing-plans
obra/superpowers
Use when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available