fix: make GitHub skills shell-neutral across platforms

[jamiechicago312]

Problem

GitHub-related skills in extensions still default to POSIX/bash-specific setup patterns. That makes Windows usage awkward even when the underlying workflows are otherwise cross-platform.

Scope

• update github, github-pr-review, github-pr-reviewer, and github-repo-monitor
• remove bash-only defaults like /tmp, heredocs, mkdir -p, tar -czf, and grep/sed-style verification from the primary guidance
• prefer shell-neutral instructions, workspace-local JSON files, Python-based packaging, and guidance that adapts to the active shell
• regenerate skills/index.js

Testing

• uv run --with pytest --with requests --with openhands-sdk pytest tests/test_github_pr_review_skill.py tests/test_skills_have_readme.py tests/test_workflow_sync.py tests/test_sync_extensions.py

Closes #323.

This PR description was created by an AI agent (OpenHands) on behalf of the user.

[neubig]

@OpenHands

1. for the bigger python snippets, could we write them as scripts in the skill directory scripts folder? that would make them more immediately reusable.
2. check enyst's review comments. if it would be possible to rely on the system temporary directory consistently that should be better.

[openhands-ai]

Uh oh! There was an unexpected error starting the job :(

[jamiechicago312]

• for the bigger python snippets, could we write them as scripts in the skill directory scripts folder? that would make them more immediately reusable.
• check enyst's review comments. if it would be possible to rely on the system temporary directory consistently that should be better.

@OpenHands address these concerns by @neubig

[openhands-ai]

I'm on it! jamiechicago312 can track my progress at all-hands.dev

[jamiechicago312]

Addressed in a96c5a9:

• restored a concrete example in skills/github-pr-review/README.md
• switched the GitHub review/GraphQL temp-file guidance to system-temp placeholders
• extracted the larger GitHub automation setup snippets into reusable helper scripts under skills/github-pr-reviewer/scripts/ and skills/github-repo-monitor/scripts/
• updated the automation docs so generated tarballs/build artifacts live outside the repository

Validation: uv run --with pytest --with requests --with openhands-sdk pytest -q tests/test_github_pr_review_skill.py tests/test_skills_have_readme.py tests/test_workflow_sync.py tests/test_sync_extensions.py

This comment was created by an AI agent (OpenHands) on behalf of the user.

[jamiechicago312]

OK openhands resolved concerns from @neubig and @enyst 🙏 tysm

[all-hands-bot]

👋 This PR needs a couple of things fixed before OpenHands can review it:

• the PR description's HUMAN: section needs at least 20 characters describing what you tested, not just the template placeholder

Push an update once this is addressed and this check re-runs automatically.

This is an automated check - no AI was used to generate this comment.

[all-hands-bot]

👋 This PR needs a couple of things fixed before OpenHands can review it:

• the PR description's HUMAN: section needs at least 20 characters describing what you tested, not just the template placeholder

Push an update once this is addressed and this check re-runs automatically.

This is an automated check - no AI was used to generate this comment.