(feat) adds dlthub agents #15

	name: Claude Code Review

	on:
	pull_request:
	branches: [master]
	types: [opened, synchronize]

	jobs:
	claude-review:
	# Skip PRs opened by bots to avoid noisy / self-review loops.
	if: >-
	github.event.pull_request.user.type != 'Bot'
	runs-on: ubuntu-latest
	permissions:
	contents: read
	pull-requests: write
	issues: read
	id-token: write
	steps:
	- name: Checkout repository
	uses: actions/checkout@v4
	with:
	fetch-depth: 1

	- name: Run Claude Code Review
	uses: anthropics/claude-code-action@30544b674398ee15c84819bd87caf8a87e8c7b55 # v1.0.154
	with:
	anthropic_api_key: ${{ secrets.ANTHROPIC_PLATFORM_KEY }}
	prompt: \|
	REPO: ${{ github.repository }}
	PR NUMBER: ${{ github.event.pull_request.number }}

	Review this pull request. This repository is the dltHub AI Workbench: a
	collection of toolkits (markdown workflows + Python tools) that give AI coding
	assistants step-by-step guidance for building dlt data pipelines.

	Follow the review conventions in the repository's CLAUDE.md and REVIEW.md.
	Focus on:
	- Correctness and clarity of toolkit instructions and Python tools
	- Consistency with existing toolkit structure and naming
	- Broken links, wrong paths, and stale references
	- Anything that would lint (see the Makefile `lint` target)

	Provide concise, actionable feedback. Post your review as a single PR comment
	using `gh pr comment`. If the change looks good, say so briefly rather than
	inventing issues.
	claude_args: \|
	--allowed-tools "Read,Grep,Glob,Bash(gh pr view:),Bash(gh pr diff:),Bash(gh pr comment:*),Bash(make lint)"

Navigation Menu