TechTide Harness Kit

The enterprise skill platform for AI coding agents.
391 production skills. 348 specialist agents. 6 cloud providers. 7 harnesses. One repo.

Get Started · Harnesses · Cloud Coverage · Trust · Anatomy · Full Catalog · Contribute

Not an awesome-list. Every skill ships with YAML-frontmatter tool scoping, inline assessment questions, safety checklists with blast-radius checks, and 9 JSON Schema contracts enforcing shape at CI time. Built for teams that answer to auditors, not just developers.

 Works With

Anthropic Claude Code Native plugin marketplace. One command installs 348 agents with full trust metadata. /plugin marketplace add TechTideOhio/techtide-harness-kit Install guide →

Cursor AI IDE Plugin directory integration. Clone once, get schema-validated skills across every provider lane. git clone TechTideOhio/ techtide-harness-kit # Add as plugin directory Install guide →

OpenAI Codex Marketplace plugin with dual-plugin architecture. Full agent catalog plus MCP trust references. codex plugin marketplace add TechTideOhio/techtide-harness-kit Install guide →

Lovable One ZIP per skill. Generate archives, upload to workspace. Vibe-coding with enterprise guardrails. npm run lovable:write # Upload ZIPs to Lovable workspace Import guide →

More harnesses: GitHub Copilot, Gemini CLI, Kiro

Harness	Install	Format
GitHub Copilot	copilot plugin marketplace add TechTideOhio/techtide-harness-kit	.github/plugin/ manifest
Gemini CLI	npx thk-export-agents --platform gemini --all --repo .	Workspace skill adapters
Kiro	Add powers/techtide-* directories in Kiro panel	Powers packages

Full comparison and install paths: docs/integrations/installation-guide.md

 Get Started

npm install @techtide/harness-kit

# Validate everything (25+ gates, no secrets required)
npm run validate

# Export agents for your harness
npx thk-export-agents --platform claude-code --provider aws --repo .
npx thk-export-agents --platform codex --role cloud-security-engineer --repo .
npx thk-export-agents --platform gemini --all --repo .

 By The Numbers

391	348	6	33
Production skills	Specialist agents	Cloud providers	Provider lanes
9	25+	7	12
JSON Schema contracts	Validation gates	Harness adapters	Core agent skills

 Cloud Coverage

Every cloud has deep, provider-specific skills with inline assessment questions and validation checklists -- not generic templates with names swapped.

Major Cloud Providers Provider Skills Highlights GCP 51 AlloyDB AI, Anthos, Cloud Run, GKE, WAF (50+ questions) AWS 47 WAF pillars, Bedrock security, Cost Explorer, IAM Alibaba 43 ACK, AnalyticDB, China compliance, ActionTrail Huawei 43 GaussDB HA, CCE, Cloud Eye, CBR backup OCI 41 Autonomous DB, Cloud Guard, Full Stack DR Azure 36 AI Foundry, AKS, App Service, Entra ID, WAF

Infrastructure & Governance Lane Skills Highlights TechTide 24 Production audit, agent autopsy, MCP trust Marketing 14 Ad fairness (FHA/ECOA/EU AI Act), data min. Core 12 Context mgmt, security, prompt hardening NVIDIA 12 CUDA kernels, TensorRT-LLM, Triton, NGC Kubernetes 10 External Secrets, RBAC, admission control CNCF 11 Argo, Cilium, Falco, Istio, Kyverno, Velero FinOps 7 Cross-cloud cost governance, showback European 30 Hetzner, Contabo, IONOS, OVHcloud, Scaleway

 Skill Anatomy

Every skill follows a consistent, agent-optimized structure with least-privilege tool scoping:

skills/<provider>/<skill-id>/
  SKILL.md            # YAML frontmatter + workflow + assessment questions + checklist
  metadata.json       # Schema-validated: id, version, harnesses, official_docs
  references/         # Safety checklists, official sources, extended guidance

Frontmatter declares tool scope --- name: techtide-aws-waf-reliability-review description: "Review AWS workload reliability..." allowed-tools: Read Grep Glob metadata: author: "github: TechTide" version: "0.1.0" category: resilience --- No Write. No Bash. No network. Least-privilege by default.

Inline assessment questions + checklists ## Assessment Question Bank ### Compute and Autoscaling 1. Are all stateless tiers behind autoscaling? 2. Is scale-in protection set for in-flight? ## Validation Checklist - [ ] Multi-AZ for all production compute - [ ] Autoscaling configured for stateless - [ ] Backup restore tested in last 30 days - [ ] Circuit breakers on external calls

 What Ships

Surface	Purpose
skills/	Portable task workflows with frontmatter, assessment questions, and guardrails
agents/	Specialist roles with harness-specific adapters
rules/	Harness-specific operating guidance
mcp/	Trusted MCP server integration references
catalog/	Machine-readable indexes, trust metadata, roles, and integrity hashes
schemas/	9 JSON Schema contracts enforcing skill and agent shape
powers/	Kiro Power packages
plugins/	Codex and Claude Code plugin packages

 Enterprise Trust

Security & governance layers Layer What it does Schema validation 9 JSON Schemas enforce skill shape and trust fields Tool scoping allowed-tools in frontmatter -- least-privilege Safety checklists Blast radius, rollback paths, data impact Approval gates Read-first, approval-gated, target-confirmed Trust metadata Tool scopes, data classes, network posture Provenance npm provenance + SHA asset integrity hashes CodeQL Automated security scanning on every push OpenSSF Scorecard Supply chain security posture

Trust documentation Document Purpose TRUST.md Trust posture and verification model DATA-HANDLING.md Data handling and privacy practices PROMPT-INJECTION.md Prompt injection defenses CONTROL-MAPPING.md Compliance control mapping SECURITY.md Vulnerability reporting EVALS.md Evaluation and validation summary

 vs Alternatives

Feature	TechTide Harness Kit	Awesome lists	Toolkit repos
Production skills	391	Links only	35-184
Cloud providers	6	0	1-2
Schema validation	9 schemas, 25+ gates	None	None
Trust metadata	Per-skill scopes + approval gates	None	None
Safety checklists	Blast radius + rollback	None	None
Multi-harness	7 adapters	1	1-2
Assessment questions	Provider-specific banks	None	None
Enterprise governance	OpenSSF, CodeQL, provenance	None	None
European cloud	5 providers, 30 skills	None	None
NVIDIA / GPU	CUDA, TensorRT, Triton	None	None

 Common Commands

# Validate everything (25+ gates)
npm run validate

# Check trust + proof layer
npm run trust:check && npm run proof-layer:check

# Export agents
npm run agents:export -- --list
npm run agents:export -- --platform claude-code --provider aws --repo .
npm run agents:export -- --platform codex --all --repo .

# Regenerate artifacts
npm run proof-layer:write && npm run plugin-manifest:write

 Documentation

Topic	Link
Installation guide	docs/integrations/installation-guide.md
Harness compatibility	docs/compatibility.md
Cross-harness skills	docs/cross-harness-skills.md
Marketplace model	docs/marketplace-model.md
Quality bar	docs/quality-bar.md
Taxonomy	docs/taxonomy.md
Lovable skill imports	docs/integrations/lovable-skills.md
External skill research	docs/external-skill-research.md

 Contributing

Contributions should be evidence-backed, source-grounded, and safe by default.

1. Read CONTRIBUTING.md
2. Run npm run validate before opening a PR
3. Every skill needs: SKILL.md with frontmatter, metadata.json matching schema, and official doc references

Report vulnerabilities through SECURITY.md. Do not open public issues containing exploit details, real credentials, customer data, or internal system identifiers.

---

Built by TechTide AI  ·  Alex Cinovoj  ·  Columbus, Ohio

Star this repo if it helps your team ship safer agent workflows.