Zenable - AI Guardrails That Clean Up Sloppy AI Code

# Zenable - AI-Native SDLC Governance & Guardrails > Production-ready tooling AI-accelerated development: correctness, conformance, observability in the SDLC. _Last-Updated: 2025-10-31 Canonical domains: https://zenable.io | https://www.zenable.app/docs | https://www.zenable.app_ ## Overview - How It Works Zenable runs where your developers already work: inside IDEs, through MCP servers that extend AI assistants, in cloud runtimes alongside kubernetes or CI/CD pipelines, and connected to ticketing systems and pull request reviews. These integrations provide seamless guardrails without disrupting existing workflows. 1. Spec-Driven Development (Guardrails) - Ensure speed and quality by enforcing conformance and custom requirements directly in the SDLC. Provide just enough context, only when it matters, to keep AI-accelerated workflows efficient and precise. 2. Continuous Improvement - Automatically measure, learn, and refine prompts, context, and guardrails to reduce token waste, keep integrations healthy, and adopt new IDE features for throughput and reliability. 3. Evidence and Observability - Instrument the SDLC with metrics, telemetry, and audit-ready evidence. Turn usage data into actionable reports that feed SDD and the improvement loop, doubling as governance metrics, compliance artifacts/assurance, and internal performance KPIs. ## Getting Started (MCP) To configure make sure `uv` is installed, and then run `uvx zenable-mcp install <foo>` where `<foo>` is the IDE name, for instance claude, codex, vscode, or many others. Use `uvx zenable-mcp install --help` to get a full list of options. Ultimately this will: - Configure streaming HTTP to **https://mcp.zenable.app** - Leverage the **`conformance_check`** tool - Use **OAuth** for auth, in line with the latest **MCP specification (2025-06-18)** - **Learn more:** https://www.zenable.app/docs/integrations/mcp/getting-started?utm_source=llms-txt&utm_medium=ai ## Getting Started (GitHub) - Go to **https://z