Help us improve
Share bugs, ideas, or general feedback.
Share bugs, ideas, or general feedback.
Share bugs, ideas, or general feedback.
By Camilool8
Curated data engineering pack with four sub-domains — analytics-engineering, data-analyst-notebook, llm-app, and ml-pipeline — wiring domain skills, read-only architect/auditor agents, warehouse read-only and sample-then-scale guards, provenance audit logging, and eval-separation discipline.
npx claudepluginhub camilool8/harness-engineering-templates --plugin harness-dataDesigns the layer cake (staging → marts → semantic layer); drafts contracts and unit tests before models. Use before any dbt model implementation.
Reviews rendered charts against the canonical sins list. Use PostToolUse on plt.savefig, fig.write_html, or matplotlib show. Different model family from the generator agent (cross-family judge).
Writes contracts before models; refuses model PRs that break an existing contract without a migration note. Use before any new staging+ model.
Emits a data hash for every input parquet / arrow / DuckDB snapshot used in a run. Refuses commits that change a model artifact without a recorded data hash. Use whenever a new dataset is introduced or an existing one changes.
Emits a dataset card structured for NIST AI RMF Map and EU AI Act Annex IV. Use whenever a new training, eval, or source dataset is introduced.
Google BigQuery remote MCP conventions (preview) — GCP Workload Identity Federation auth never a static service-account key, a read-only bigquery.jobUser + dataViewer role, mandatory partition pruning, and in-place query streaming. Use when querying BigQuery, wiring the BigQuery MCP, or setting up read-only warehouse access on GCP.
Databricks Managed MCP conventions (Public Preview) — Unity Catalog ACLs as the upstream gate, OAuth only never a personal access token, Genie spaces treated as downstream lineage consumers, and Mosaic AI model serving published by humans/CI not the agent. Use when querying Databricks, wiring the Databricks MCP, or working with Unity Catalog.
dbt-core conventions — contract-first models with contract.enforced true, mandatory unit tests, a semantic layer as the single metric source of truth, the dbt remote MCP as the governed agent surface, and auto-activating dbt-agent-skills. Use when building dbt models, contracts, unit tests, or semantic-layer metrics, or wiring the dbt MCP.
DuckDB + MotherDuck conventions — local-first SQL over Parquet/CSV/JSON with no warehouse round-trip, MotherDuck OAuth (never a static token), sample-then-scale even locally, and DuckDB as a join engine for Polars via pl.SQLContext. Use when querying local data files or using DuckDB/MotherDuck. The duckdb MCP server ships with this plugin.
UK AISI Inspect AI eval conventions — the dataset/solver/scorer three-piece shape, Docker-sandboxed task isolation, the 200+ pre-built inspect_evals starter tasks, and wrapping Claude Code/Codex/Gemini CLI as the agent under test. Use when authoring or running Inspect AI evals, sandbox-isolated agentic evals, or adding a custom eval task.
Executes bash commands
Hook triggers when Bash tool is used
Modifies files
Hook triggers on file write and edit operations
Share bugs, ideas, or general feedback.
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge.
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge.
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
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 skill pack with 66 specialized skills for full-stack developers: 12 language experts (Python, TypeScript, Go, Rust, C++, Swift, Kotlin, C#, PHP, Java, SQL, JavaScript), 10 backend frameworks, 6 frontend/mobile, plus infrastructure, DevOps, security, and testing. Features progressive disclosure architecture for 50% faster loading.
Comprehensive startup business analysis with market sizing (TAM/SAM/SOM), financial modeling, team planning, and strategic research
v9.42.3 — Patch release for cursor-agent smoke checks in untrusted workspaces. Run /octo:setup.
Intelligent draw.io diagramming plugin with AI-powered diagram generation, multi-platform embedding (GitHub, Confluence, Azure DevOps, Notion, Teams, Harness), conditional formatting, live data binding, and MCP server integration for programmatic diagram creation and management.
Comprehensive .NET development skills for modern C#, ASP.NET, MAUI, Blazor, Aspire, EF Core, Native AOT, testing, security, performance optimization, CI/CD, and cloud-native applications
Curated DevOps engineering pack with four sub-domains — cicd-platform, infrastructure, kubernetes-platform, and observability-sre — wiring domain skills, read-only architect/auditor agents, propose-never-apply IaC and Kubernetes safety hooks, and a Sigstore supply-chain posture.
Four non-negotiable safety hooks (command-guard, secret-scan, verify-gate, audit-log) plus cross-cutting engineering modules as auto-loading skills. Every Harness domain pack depends on this.
Curated mobile engineering pack with four sub-domains — flutter-app, native-android, native-ios, and react-native-expo — wiring domain skills, read-only compliance/UX-critic agents, a simulator/emulator-in-the-loop verification skill, store-compliance gates, and an OAuth-first credential posture.
Curated web engineering pack with five sub-domains — api-service, design-system, distributed-backend, frontend-app, and fullstack-app — wiring domain skills, read-only architect/auditor agents, accessibility-first verification, and Core Web Vitals gates.
Uses power tools
Uses Bash, Write, or Edit tools
Uses power tools
Uses Bash, Write, or Edit tools
Share bugs, ideas, or general feedback.
A library of opinionated Claude Code harness templates for software engineering across four curated domains — web, data, DevOps, and mobile.
The premise (per Birgitta Böckeler, Anthropic engineering, METR, and the practitioner consensus of 2025–2026):
Agent = Model + Harness. Templates that lean on the model's good judgement fail their first incident. The harness — not the agent — is the contract.
The full reasoning lives in docs/explanation/why-harness.md.
Install from inside Claude Code — add the marketplace, then install the pack for your domain:
/plugin marketplace add Camilool8/harness-engineering-templates
/plugin install harness-web@harness-engineering # or harness-data / harness-devops / harness-mobile
/harness-web:init # pick a sub-domain
Installing any domain pack pulls in harness-base automatically (the four non-negotiable safety hooks). Full walk-through: docs/tutorials/getting-started.md. Plugin catalog and the HARNESS.toml opt-in flags: docs/reference/plugins.md.
Prefer committed
.claude/artifacts in your repo (audit-heavy or regulated teams)? The bash assembler is still supported as an eject path — seedocs/reference/eject.md.
Five installable plugins — a shared harness-base plus four curated domain packs (harness-web, harness-data, harness-devops, harness-mobile):
.claude/HARNESS.toml; memory, progress-tracking, methodology, and orchestration guidance ship as auto-loading skills.permissions block; see docs/reference/recommended-permissions.md for an opt-in starting point.Every default is the 2026 practitioner consensus.
| You want to… | Read |
|---|---|
| Install a pack for the first time | docs/tutorials/getting-started.md |
Browse the plugin catalog + HARNESS.toml flags | docs/reference/plugins.md |
| Pick a domain pack and sub-domain | docs/how-to/pick-a-recipe.md |
| Use the bash assembler (eject path) | docs/reference/eject.md |
| Understand the design | docs/explanation/why-harness.md |
| Contribute a module, addon, or sub-domain | docs/how-to/ and CONTRIBUTING.md |
The full documentation index, organised by Diátaxis quadrant: docs/README.md.
Three deep-reference essays (the why behind the design):
docs/HARNESS_ENGINEERING.md — foundations, the Claude Code primitive reference, per-domain templates, cross-cutting concerns, universal anti-patterns.docs/METHODOLOGIES.md — SDD, TDD, BDD, ATDD, DDD, Agile, Waterfall, Lean, DevOps/SRE, Eval-Driven Development — each adapted to AI harnesses.docs/AGENT_ROLES.md — single-agent baseline, multi-agent topologies, the canonical role catalog, sub-agent design principles.docs/ user-facing documentation (start here)
tutorials/ learning-oriented walk-throughs
how-to/ problem-oriented recipes
reference/ schemas, CLI, catalogs, glossary
explanation/ understanding-oriented design notes
HARNESS_ENGINEERING.md deep reference
METHODOLOGIES.md deep reference
AGENT_ROLES.md deep reference