Stats
Actions
Tags
From dev-pipeline
Mid-review entry point: fetch PR comments and CI failures, address them, validate, push. Use when jumping into an existing PR. /dev handles this automatically.
How this skill is triggered — by the user, by Claude, or both
Slash command
/dev-pipeline:address-prThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
For when you jump into a session with an existing PR that needs attention. Fetches comments and CI failures, addresses them, validates, and pushes. `/dev` does this automatically in its review loop.
For when you jump into a session with an existing PR that needs attention. Fetches comments and CI failures, addresses them, validates, and pushes. /dev does this automatically in its review loop.
Commit and push after every logical fix. Don't batch all review fixes into one commit. Formatting/lint fixes from dev_checks get their own commit.
Run pwd. Must contain .claude/worktrees/.
If not: STOP with error.
gh pr view --json number,url,title
If no PR exists for this branch, tell the user and stop.
Review comments:
gh repo view --json nameWithOwner --jq .nameWithOwner
gh pr view <number> --json reviews --jq '.reviews[] | select(.state != "APPROVED") | "\(.author.login) (\(.state)): \(.body)"'
gh api repos/<owner>/<repo>/pulls/<number>/comments --jq '.[] | "\(.path):\(.line // .original_line) — \(.user.login): \(.body)"'
CI status:
gh pr checks <number>
If checks failed, fetch logs:
gh run view <run_id> --log-failed
Present everything clearly — comments and failures together.
Work through each item:
Run full local validation:
uv run pytest tests/unit_tests/ -v./scripts/dev_checks.shIf dev_checks fixes formatting/lint, commit and push those separately (style: fix lint/formatting).
Run git status to verify nothing is uncommitted. Everything should already be pushed from step 4.
gh pr view <number> --json reviewDecision,mergeStateStatus
Report:
The human decides when to merge.
npx claudepluginhub luthienresearch/internal-utils --plugin dev-pipelineProvides behavioral guidelines to reduce common LLM coding mistakes, focusing on simplicity, surgical changes, assumption surfacing, and verifiable success criteria.
Searches, retrieves, and installs Agent Skills from prompts.chat registry using MCP tools like search_skills and get_skill. Activates for finding skills, browsing catalogs, or extending Claude.
Creates, edits, and optimizes skills for Claude Code, including drafting, evaluating with test prompts, iterating on performance, and improving skill descriptions for better triggering accuracy.