用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
直接命令不会经过审查 Prompt；运行前请先检查来源。
npx skills add https://github.com/polarsource/polar --skill interview-task命令会保持在同一行。复制前请横向滚动并检查完整内容。
想先保存到本地？可下载 SkillsMP 当前能够提供的文件。
Integrate Polar billing in server-side Python applications using the versioned Polar and PolarAsync clients. Use when implementing Polar customers and external IDs, checkout or customer portal sessions, customer-state entitlement checks, webhook validation and processing, API error handling, usage-event ingestion and metered billing, or migrating an application from the old Polar Python SDK.
Local development environment management for Polar using Docker
Evidence-capture protocol for verifying web/dashboard/backoffice/checkout changes in the Polar local stack by driving the real UI with Playwright. Auto-discovered by the built-in /verify skill; can also be invoked directly. Brings up the Docker stack, logs in via the real email-OTP flow, and exercises flows end-to-end (including a live Stripe checkout) capturing screenshots as evidence.
正在显示 SKILL.md
基于 SOC 职业分类
| name | interview-task |
| description | Prepare an interview task for a candidate, as part of our hiring process. |
| user-invocable | true |
| allowed-tools | Bash(gh:*) Bash(git:*) |
This skill is a complete cookbook for creating an interview task for a candidate. It includes steps to create the necessary materials. The idea is to create a private repository for the candidate, and have an LLM agent draft a PR for a given task. The candidate will then be asked to review the PR, and provide feedback on it. This will allow us to evaluate the candidate's ability to understand the codebase, provide constructive feedback and good insights to get the task to the finish line.
If not provided in the invocation prompt, you should ask the user for the following information:
Run the script provided in this skill .claude/skills/interview-task/scripts/fork-repository with the candidate's name as an argument. This will create a private repository for the candidate.
The name of the candidate should be slugified (lowercase, ASCII, spaces replaced with hyphens) to be used in the repository name.
Example
.claude/skills/interview-task/scripts/fork-repository francois-voron
Run the script provided in this skill .claude/skills/interview-task/scripts/setup-remote with the candidate's slug and the branch name as arguments. This will add a remote pointing to the candidate's fork, fetch it, create the branch, and push it to the remote. The branch name should be descriptive of the task.
Example
.claude/skills/interview-task/scripts/setup-remote francois-voron add-new-feature
Use an LLM agent to draft a PR in the candidate's repository. The PR should be based on the prompt provided by the user. The LLM agent should create one or more commits that implement the changes described in the prompt.
Once the changes have been made, push the changes and create a pull request in the candidate's repository. The PR should be titled and described in a way that clearly communicates the changes that were made.
git push && gh pr create --title "Add new feature" --body "This PR adds a new feature that allows users to do X, Y, and Z."
Post a comment to this PR following the template below, asking the candidate to review the PR and provide feedback on it.
Hi <candidate>,
We have created this PR as part of our interview process. We would like you to review the changes and provide feedback on it. Please take your time to understand the codebase, and provide constructive feedback and good insights to get the task to the finish line.
## Expected Outcome
- A clear understanding of the changes made in the PR
- Constructive feedback on the implementation, including any potential issues or improvements that could be made
- A discussion on how to get the task to the finish line, including any additional changes that may be needed to complete the task
We **don't** expect you to actually implement the changes.
## Next Steps
Post your feedback as a comment on this PR, and we will review it together during our next interview session.
Command to post the comment:
gh pr comment <pr-number> --body '<body>'