By rjskene
Claude Pipeline — CI workflow for automating code updates through GitHub issues (classify → plan → review → execute → PR review).
Read-only hygiene pass over the open-issue set; surfaces likely duplicates, tracker fits, missing labels, and merged-PR supersession candidates. Usage: /pipeline:analyze-issues
Run a coordinated-leg campaign over the approved slate — the OUTER loop that partitions work into ordered, cap-bounded legs and sequences them so the global rate-limit budget is spent in bounded batches. Equivalent standalone entry to `/pipeline:fullsend --campaign`. Usage: /pipeline:campaign [issue_numbers...] [--manual-merge] [--spawn] [--max-bc=N] [--max-ad=N]
Triage a pipeline issue — reads title/body/labels/comments, recommends PATH A (docs-only), B (standard), C (multi-task), or D (quick-fix), and applies the `docs-only` / `multi-task` / `quick-fix` label directly. Posts a `## Classification` comment. Usage: /pipeline:classify-issue <issue_number>
Brainstorm mode — discuss code changes freely, then push actionable items as GitHub issues instead of implementing them directly.
Validate consumer install state — read-only audit of pipeline.config, gh auth, GitHub labels, plugin registration, residual subtree artifacts, and base branch. `--fix labels` seeds the canonical pipeline labels idempotently; `--fix config` reconciles new PIPELINE_* knobs into the host pipeline.config (append-only, never overwriting host values). Usage: /pipeline:doctor [--fix labels | --fix config]
Matches all tools
Hooks run on every tool call, not just specific ones
Executes bash commands
Hook triggers when Bash tool is used
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
Modifies files
Hook triggers on file write and edit operations
Modifies files
Hook triggers on file write and edit operations
Uses power tools
Uses Bash, Write, or Edit tools
Uses power tools
Uses Bash, Write, or Edit tools
Harness and orchestrator for GitHub-issue-driven CI workflows on Claude Code
Pipeline manages GitHub issues end-to-end through automated stages, with explicit human checkpoints for plan approval and PR merge.
create → classify → plan → eval → execute → eval-pr → merge
Full process maps in docs/process-maps.md.
| Command | When to use |
|---|---|
/pipeline:status | Interactive — check pipeline status, see what's ready, advance the next stage (/pipeline:run remains as a deprecated alias) |
/pipeline:fullsend [N ...] | Autonomous end-to-end run for one or many issues (classify → plan → evaluate-plan → execute → evaluate-pr → greenlight-merge); --campaign partitions the slate into ordered per-path legs (expensive B/C vs cheap A/D) for cost-bounded autonomous runs |
/pipeline:campaign [N ...] | Standalone entry to the coordinated-leg campaign loop — equivalent to /pipeline:fullsend --campaign (same machinery); --max-bc=N / --max-ad=N override the per-leg caps |
/pipeline:analyze-issues | read-only hygiene pass — duplicate / tracker-fit / missing-label / supersession detection |
/pipeline:init | Bootstrap a fresh project — preflight deps / detect repo+branch / generate gitignored pipeline.config / seed labels / doctor audit |
Full command catalogue (every skill, all flags, interaction surfaces): see docs/skills-api.md.
tdd-implementer fan-out) / D (quick-fix); the classifier sends each issue to the cheapest path that fits. See docs/process-maps.md./usage endpoint gates dispatch against the live 5h/7d budget; fail-open with an opt-out kill switch, so a stale read never blocks a run. See docs/usage-gate.md.pause-5h) or hard-halts (halt-7d) instead of burning into a rate-limit wall (7d wins over 5h). See docs/usage-gate.md.PIPELINE_PATH_{B,D}_MODEL_EXECUTE, gated by the low-blast eligibility predicate); the Opus pr-eval backstop is mandatory and never tier-dropped. See docs/cost-architecture.md and docs/analysis/model-downsampling.md./pipeline:tokenomics renders per-bucket / per-stage / per-structure cost and latency with the B→D breakeven over the gated agent-costs.jsonl log (dogfood-only). See docs/observability.md and docs/tokenomics/README.md.[split-role-red] commit), a cheaper implementer greens it, and scripts/split-role-gate.sh asserts the locked red suite was never modified/deleted and is green at HEAD before auto-merge. See docs/split-role-tdd.md./pipeline:hotfix is the in-session emergency lane that bypasses the lifecycle gates, and every issue executes in its own git worktree so the main workspace stays clean. See docs/architecture.md and skills/hotfix/SKILL.md.npx claudepluginhub rjskene/pipeline --plugin pipelineHarness-native ECC operator layer - 67 agents, 271 skills, 92 legacy command shims, reusable hooks, rules, selective install profiles, and production-ready workflows for Claude Code, Codex, OpenCode, Cursor, and related agent harnesses
Upstash Context7 MCP server for up-to-date documentation lookup. Pull version-specific documentation and code examples directly from source repositories into your LLM context.
Comprehensive feature development workflow with specialized agents for codebase exploration, architecture design, and quality review
A growing collection of Claude-compatible academic workflow bundles. Covers scientific figures, manuscript writing and polishing, reviewer assessment, citation retrieval, data availability, paper reading, literature search, response letters, paper-to-PPTX conversion, and evidence-grounded Chinese invention patent drafting. Rules are organized as reusable skill folders with explicit workflows and quality checks.
Core skills library for Claude Code: TDD, debugging, collaboration patterns, and proven techniques
Reliable automation, in-depth debugging, and performance analysis in Chrome using Chrome DevTools and Puppeteer