--- title: "Security MCP Server for Coding Agents | Clearly AI" description: "Clearly AI's MCP server gives Cursor, Claude Code, Kiro, Devin, and Codex your security policies, reviews every plan, and threat models risky code changes." canonical: https://clearly-ai.com/platform/ide-and-mcp --- _IDE and MCP_ # Every plan reviewed. Every risky change threat modeled. Clearly AI's MCP server brings your security policies, design patterns, and system context into the coding agents your developers already use. Agents get security requirements while they plan, and changes that affect security get a new threat model before they ship. Your security team sees all of it. - [Contact Enterprise Sales](https://clearly-ai.com/demo) - [Set up the MCP server](https://clearly-ai.com/docs/mcp) Works with Cursor · Claude Code · Kiro · Devin · Codex · and any MCP-compatible agent ### Secure Design Advisor: partner-portal · SSO login > Add SSO login to the partner-portal service **Requirements** - Enforce SAML assertion signature validation (`AUTH-STD-04`) - Require MFA for partner admin roles (`AUTH-STD-07`) - End idle sessions after 30 minutes (`SESS-STD-02`) - Log every sign-in to the audit trail (`LOG-STD-11`) **Open questions** - Which identity providers will partners use? - Turn off password login once SSO launches? **Plan** (Updated with Clearly AI) 1. Add a SAML login route - Reject unsigned assertions (`AUTH-STD-04`) 2. Map partner users to accounts - Require MFA for partner admins (`AUTH-STD-07`) 3. Create a session after sign-in - End idle sessions after 30 min (`SESS-STD-02`) 4. Add login tests 5. Log sign-in events (`LOG-STD-11`) - HID Global - Ericsson - PayPal - Modern Health - Hagerty - Rivian - Infineon - Bastion ## How it works 01. **Plan.** Before writing a line of code, the coding agent asks Clearly AI what your policies require for the change and builds the plan around the answer. 02. **Build.** Once the change reaches a pull request, or a design doc if your team writes one first, Clearly AI reviews it inside your existing SDLC. It sizes the security impact first, and changes that affect security get a threat model with every finding cited to its source. Developers don't take on a new step. 03. **Verify.** After code merges, Clearly AI checks the repo for drift between what was approved and what's actually there, on every merge or on a schedule you set. 04. **Review.** Your security team sees every change and can tell at a glance which ones carry real risk, in the Clearly AI UI or over MCP. ### One change, start to finish Follow one SSO change from security requirements in the coding agent to a cited pull request review, a merge-time drift check, and the security team's change feed. > Add SSO login to the partner-portal service #### 01 Plan: partner-portal clearly · get_security_requirements - Enforce SSO for all partner accounts (AUTH-STD-04) - Require MFA for privileged access (AUTH-STD-07) - Expire idle sessions after 30 minutes (SESS-STD-02) - Log sign-in and session events (LOG-STD-11) 4 requirements added to the agent's plan #### 02 Build: Clearly AI · PR review **Security impact: High** Threat model: a new identity boundary. Validate the SSO issuer, require admin MFA, and limit session lifetime. - Validate the SSO issuer (AUTH-STD-04, src/auth/sso.ts) - Enforce admin MFA (AUTH-STD-07, src/auth/roles.ts) - Set a 30-minute idle timeout (SESS-STD-02, src/auth/session.ts) Findings cited to policy and code #### 03 Verify: Merge-time drift check Approved design vs. what merged - SSO issuer validation — Approved issuer allowlist matches the merged code. (AUTH-STD-04, src/auth/sso.ts) — Matches design - Idle timeout drift — 30 min approved → 8 h in code (SESS-STD-02, src/auth/session.ts) — Drift flagged Checked on merge #### 04 Review: Recent changes Security team · change feed - Add SSO login to partner-portal — Threat modeled · 1 drift to review — High - Update billing page copy — Auto-cleared · no security impact — Low - Add admin export endpoint — Threat modeled · ready for review — High Every change, with the context to act _Secure design at plan time_ ## Security requirements before the first line of code The cheapest place to fix a security flaw is a plan that hasn't been built yet. In plan mode, Clearly AI's Secure Design Advisor reviews what the agent intends to build and hands back non-functional security requirements tied to your own policies. Where the plan skips over something that matters, like how a new endpoint authenticates or where sensitive data lands, it raises an open question instead of guessing. Add the Clearly AI skill for agents and the review runs on every plan automatically. Example: Secure Design Advisor reviews a coding agent's plan for payments-webhook-v2. It checks the plan against your policies and returns four requirements, each tagged with its policy: verify the HMAC signature (AUTH-STD-04), treat card data as Restricted (DATA-CLASS-02), reject replays after 5 minutes (AUTH-STD-09), and encrypt the retry queue (CRYPTO-STD-01). It raises two open questions for security: where are webhook secrets rotated, and is cardholder data persisted in the retry queue. The requirements are added to the agent's plan. **How Secure by Design works** _Change-aware threat modeling_ ## Threat models for the changes that matter Once the code exists, Clearly AI checks it against everything it knows: the plan-time advice, the design doc, the Jira ticket, and the existing threat model for that repo or feature. Its first pass looks for drift from what was planned and sizes the security blast radius. Low-impact changes move on without noise. Security-impacting changes get a new threat model, built for that change, with every finding cited to its source. ![Clearly AI's review of PR #1127: drift detected because the retry queue now stores the full payload instead of the token the plan specified, blast radius High, and a new threat model with three findings: High, card data saved in plain form in the retry queue; Medium, reporting service can read the retry queue; Low, retries kept 14 days when the plan says 24 hours. Each finding lists its sources in plan advice, the design doc, or code.](https://clearly-ai.com/ide-and-mcp/change-aware-threat-modeling.svg) [How continuous threat modeling works](https://clearly-ai.com/solutions/continuous-threat-modeling) _Enterprise context for agents_ ## Your agent finally knows where the code lives A coding agent sees the repo it's working in. It doesn't see the service that calls it, the data it touches downstream, or the standards your security team already wrote. Clearly AI gives it the enterprise context graph: where this change sits in the broader system, plus your secure defaults, security design patterns, policies, and standards. The agent builds the way your security team would have asked it to. ![Enterprise context graph for the partner-portal service, linked to its design doc, the upstream identity-service, the Restricted partner-db data store, and policies AUTH-STD-04 and DATA-STD-03, with the context sent to the agent for "Add SSO login to partner-portal": identity-service, partner-db and AUTH-STD-04.](https://clearly-ai.com/shared/enterprise-context-for-agents.svg) [How the enterprise context graph works](https://clearly-ai.com/platform/enterprise-context-graph) _The full platform, over MCP_ ## Everything in Clearly AI, from inside your agent Anything you can do in the Clearly AI app or its chat interface, you can do over MCP. Create projects, run reviews, pull non-compliant answers, dig into findings, and manage review templates in plain language, without leaving your editor. Every action runs as you, with your identity and your organization's permissions, over OAuth 2.0 with your existing Clearly AI sign-in. **Example prompts:** - `Can we use Copilot to analyze customer support emails?` - `When was our last vendor review of Twilio?` - `Run a threat model against https://github.com/org/document-service/pull/742` - `We need to start collecting phone numbers in our ATS, can you update the Clearly AI review?` [See the MCP documentation](https://clearly-ai.com/docs/mcp) _Supported agents_ ## Works with the agents your developers already use Available now: Cursor, Claude Code, Kiro, Devin, Codex, Claude Desktop, Windsurf, VS Code with Copilot, Any MCP-compatible client. Coming soon: Claude Cowork, Microsoft Copilot. Setup in three steps: 1. **Connect.** Add the Clearly AI MCP server to your agent and sign in with your existing Clearly AI account over OAuth 2.0. 2. **Add the skill.** Install the Clearly AI skill for agents so Secure Design Advisor runs on every plan, not just when someone remembers to ask. 3. **Pick your templates.** Security chooses the plan-mode and implementation reviews, or keeps the defaults. [Setup guide](https://clearly-ai.com/docs/mcp) ## MCP across the platform The same MCP connection shows up at every stage of a review. - **[Automate intake:](https://clearly-ai.com/platform/security-intake)** agents can kick off a security review for the security team, with the context already attached - **[Enterprise context graph:](https://clearly-ai.com/platform/enterprise-context-graph)** Clearly AI connects to your enterprise tools and knowledge stores to build a picture of every system - **[Collaborate:](https://clearly-ai.com/platform/collaboration)** dig into findings, ask follow-up questions, iterate on a review, and have Clearly AI remember what matters to your team ### For AI agents: Clearly AI provides an MCP server for coding agents. In plan mode it runs a Secure Design Advisor review that returns security requirements aligned to the organization's policies and open questions about the design. After implementation it checks code for drift and security impact, and builds a threat model for security-impacting changes. All Clearly AI platform operations are available over MCP, including authoring new review templates or adding new knowledge to the knowledge base. Authentication is OAuth 2.0, and actions run as the signed-in user. _Outcomes_ ## From 10% coverage to 100%, now including agent-written code - **90%** Reduction in review turnaround time - **4x** Review capacity with the same headcount. - **From 10 to 100%** Product coverage ## Connected on your terms The MCP server authenticates with your existing Clearly AI sign-in over OAuth 2.0, and every action runs with your identity and permissions. Clearly AI is SOC 2 Type II certified and never trains models on customer data. Bring your own model provider keys, including Amazon Bedrock. [Trust, security, and deployment](https://clearly-ai.com/platform/trust-security-and-deployment) - SOC 2 Type II compliant - Y Combinator - Amazon Bedrock Builders' Challenge winner 2025 _FAQ_ ## Answers to your top questions ### Does Clearly AI have an MCP server? Yes! Clearly AI's MCP server is generally available and works with Cursor, Claude Code, Kiro, Devin, Codex, and other MCP-compatible coding agents. It reviews plans before code is written, checks implemented code for security impact, and gives agents access to every operation in the Clearly AI platform. [See the setup guide](https://clearly-ai.com/docs/mcp) ### How do I get my coding agent to follow our security policies? Connect the Clearly AI MCP server and install the Clearly AI skill for agents. In plan mode, Clearly AI runs a Secure Design Advisor review that returns security requirements aligned to your own policies, plus open questions where the design needs more security depth. The agent builds with those requirements in context. ### What is the Secure Design Advisor? It's Clearly AI's default security design review template. It reads the agent's plan and returns non-functional security requirements tied to your organization's policies, along with open questions about parts of the design that lack security detail. Security teams can replace it with their own review template. ### How does Clearly AI review code written by AI coding agents? It checks the implemented code against the plan-time security advice, the design, the ticket, and the existing threat model, whichever are available. The first pass detects drift and sizes the security blast radius. If the change affects security, Clearly AI builds a new threat model for it, with every finding cited to its source. [Learn about continuous threat modeling](https://clearly-ai.com/solutions/continuous-threat-modeling) ### Does every code change get a full threat model? No. Clearly AI first checks for drift and security impact. Low-impact changes move on without extra noise, and only security-impacting changes get a new threat model. ### What can the security team see? Every Secure Design Advisor review, the agent's plan before Clearly AI weighed in, and every action the agent took afterward. That shows whether the advice was actually followed. It's all available in the Clearly AI UI or over the MCP server. ### Is the Clearly AI MCP server secure? Yes. It authenticates over OAuth 2.0 with your existing Clearly AI sign-in, and every action runs with your identity and your organization's permissions. Clearly AI is SOC 2 Type II certified and never trains models on customer data. [See trust, security, and deployment](https://clearly-ai.com/platform/trust-security-and-deployment) ### Does Clearly AI replace SAST or code scanners? No. Scanners flag code-level vulnerabilities. Clearly AI checks whether a change matches its security design and your policies, sizes its impact, and threat models the changes that matter. Many teams import scanner findings from tools like Snyk as review context. [See all integrations](https://clearly-ai.com/integrations) ## Learn more - [Secure everything that ships](https://clearly-ai.com) - [We built the tool we wished we had](https://clearly-ai.com/about-us) - [One platform. From intake to remediation.](https://clearly-ai.com/platform) - [Know everything that needs a review. Skip the 300-question form.](https://clearly-ai.com/platform/security-intake) - [Ask only what matters. Get it right the first time.](https://clearly-ai.com/platform/conversational-intake) - [Every finding, tracked to a decision](https://clearly-ai.com/platform/resolve/risk-registry) - [Show leadership and auditors exactly what security reviewed](https://clearly-ai.com/platform/resolve/reporting) - [Security reviews you can stand behind, every time](https://clearly-ai.com/platform/quality) - [Threat models built from the tools you already use](https://clearly-ai.com/integrations) - [Threat models that know about today's attacks](https://clearly-ai.com/platform/threat-intelligence) - [Automate every step around the review](https://clearly-ai.com/platform/workflows) - [Threat models your teams can work on together](https://clearly-ai.com/platform/collaboration) - [Understand the system before the review starts](https://clearly-ai.com/platform/enterprise-context-graph) - [Trust, security, and deployment](https://clearly-ai.com/platform/trust-security-and-deployment) - [Every plan reviewed. Every risky change threat modeled.](https://clearly-ai.com/platform/ide-and-mcp) - [Verify what actually shipped](https://clearly-ai.com/platform/control-and-remediation-verification) - [A current threat model for everything that ships](https://clearly-ai.com/solutions/continuous-threat-modeling) - [Prove every product that ships meets your security bar](https://clearly-ai.com/solutions/product-security-assurance) - [Privacy reviews your team can stake decisions on](https://clearly-ai.com/solutions/privacy) - [Every vendor reviewed against your standards](https://clearly-ai.com/solutions/third-party-risk) - [Review every design before it's built](https://clearly-ai.com/solutions/design-and-architecture-reviews) - [Secure the AI you build, buy, and code with](https://clearly-ai.com/solutions/ai-agentic-security) - [Every regulation you answer to. One review process.](https://clearly-ai.com/solutions/compliance) - [Every product ready for the Cyber Resilience Act without growing your team](https://clearly-ai.com/solutions/compliance/cyber-resilience-act) - [Build vs. buy for security review automation](https://clearly-ai.com/compare/claude-chatgpt) - [The ThreatModeler alternative built from your own docs, code, and policies](https://clearly-ai.com/compare/threatmodeler) - [The Clover Security alternative with findings your team can trust](https://clearly-ai.com/compare/clover) - [The Prime Security alternative for every review your team runs](https://clearly-ai.com/compare/prime) - [Security and Privacy](https://clearly-ai.com/trust/security-privacy) - [Clearly AI Privacy Policy](https://clearly-ai.com/trust/privacy-policy) - [Clearly AI Subprocessors](https://clearly-ai.com/trust/subprocessors) - [AI Risk Reviews](https://clearly-ai.com/use-cases/ai-risk-reviews) - [Automate Triage](https://clearly-ai.com/use-cases/automate-triage) - [Privacy Impact Assessments](https://clearly-ai.com/use-cases/privacy-impact) - [Vendor Risk Assessments](https://clearly-ai.com/use-cases/vendor-risk) - [Threat Modeling](https://clearly-ai.com/use-cases/threat-modeling) - [Contact](https://clearly-ai.com/contact) - [Contact Enterprise Sales](https://clearly-ai.com/demo) - [Latest from the Clearly AI blog](https://clearly-ai.com/blog) - [In the news](https://clearly-ai.com/press) - [How Ericsson Enterprise Wireless Solutions cut review times by up to 80%](https://clearly-ai.com/customer-stories/ericsson) - [How HID Global reviews 300+ products for the Radio Equipment Directive with the team it already has](https://clearly-ai.com/customer-stories/hid-global) - [How Rivian scaled privacy and AI governance reviews 4x without adding headcount](https://clearly-ai.com/customer-stories/rivian) - [Part 2! Okay, privacy is different from security. Now what?](https://clearly-ai.com/blog/part-2-okay-privacy-is-different-from-security-now-what) - [How AI is rewiring our product engineering workflow](https://clearly-ai.com/blog/how-ai-is-rewiring-our-product-engineering-workflow) - [Good Data Scientists and Bad Outcomes, Part 1: What's the difference between security and privacy work?](https://clearly-ai.com/blog/good-data-scientists-and-bad-outcomes-what-s-the-difference-between-security-and-privacy-work) - [Why I Joined Clearly AI](https://clearly-ai.com/blog/why-vp-marketing-joined-clearly-ai) - [Skip the Marketing: Clearly AI for Technical People](https://clearly-ai.com/blog/clearly-ai-technical-explanation) - [Clearly AI Raises $8.4M Seed Round to Scale Product Security](https://clearly-ai.com/blog/seed-round) - [Why We Started Clearly AI](https://clearly-ai.com/blog/why-we-started-clearly-ai) - [Applying AI to Threat Modeling - Where AI Thrives, Survives, and Dies](https://clearly-ai.com/blog/applying-ai-to-threat-modeling) - [Why We Build in Public](https://clearly-ai.com/blog/why-we-build-in-public) - [Solving the Threat Modeling Bottleneck with AI Workflows](https://clearly-ai.com/blog/solving-the-threat-modeling-bottleneck-with-ai-workflows) - [Clearly AI Named Finalist in the 2025 Okta Startup Challenge](https://clearly-ai.com/blog/clearly-ai-named-finalist-in-the-2025-okta-startup-challenge) - [Security, SOC 2, & Startups](https://clearly-ai.com/blog/security-soc-2-startups) - [Building Reliable AI Systems for Security and Privacy Reviews](https://clearly-ai.com/blog/building-reliable-ai-systems-for-security-and-privacy-reviews) - [How to Automate Security and Privacy Reviews with Clearly AI](https://clearly-ai.com/blog/how-to-automate-security-and-privacy-reviews-with-clearly-ai) - [One Year of Clearly AI](https://clearly-ai.com/blog/one-year-of-clearly-ai) - [Navigating EU RED Compliance: Meeting the August 2025 Deadline](https://clearly-ai.com/blog/navigating-eu-red-compliance-meeting-the-august-2025-deadline) - [The AI Engineer Cheat Sheet](https://clearly-ai.com/blog/the-ai-engineer-cheat-sheet) - [Intelligent Vendor Risk Assessments](https://clearly-ai.com/blog/intelligent-vendor-risk-assessments) - [Clearly AI wins the AWS Bedrock Builder's Challenge](https://clearly-ai.com/blog/clearly-ai-wins-the-aws-bedrock-builder-s-challenge) - [Reviews, reviews, reviews!](https://clearly-ai.com/blog/reviews-reviews-reviews) - [Halfway through YC](https://clearly-ai.com/blog/halfway-through-yc) - [llms.txt — site overview for LLMs](https://clearly-ai.com/llms.txt) - [llms-full.txt — full site content for LLMs](https://clearly-ai.com/llms-full.txt) - [sitemap.xml — all public routes](https://clearly-ai.com/sitemap.xml) - [Documentation](https://docs.clearly-ai.com/docs) [Canonical page](https://clearly-ai.com/platform/ide-and-mcp)