Four Claude Code skills (code-explore, reverse-spec, smart-sdd, domain-extend) that augment spec-kit SDD workflows with cross-Feature context, 5-axis Domain Profile, and extensible domain modules
-
Updated
Mar 27, 2026 - TeX
#
clade-code
Here are 3 public repositories matching this topic...
Semantic LLM firewall for the Model Context Protocol. A local proxy that classifies every untrusted tool result, tool description and server request with an 86M encoder, pins tool manifests, tracks session taint, and blocks indirect prompt injection before it reaches your agent. Runs offline for $0. Apache-2.0. Design phase.
python proxy mcp onnx deberta anthropic llm-security agentic-ai antigravity model-context-protocol claude-code ai-firewall ai-agent-security indirect-prompt-injection clade-code rompt-injection
-
Updated
Oct 19, 2026 - HTML
A practitioner's primer on agentic development: 65 hours, zero hand-written code, one live commercial app. Method, guardrails, and working skills included.
primer ios skills multi-agent gitflow ai-agents swiftui ai-coding ai-collaboration claude-code agentic-development claude-code-cli agentic-design clade-code agentic-delivery
-
Updated
Jul 28, 2026
Add this topic to your repo
To associate your repository with the clade-code topic, visit your repo's landing page and select "manage topics."