{scd}

Security scanning for teams
shipping classic or AI-generated code

Integrity first — Your code stays in your environment

scd is an open-source SAST scanner — it finds vulnerabilities in your codebase before they reach production. Self-hosted, privacy-first, and built for development teams shipping classic and AI-generated code.

$ npm install -g @activemind/scd npmjs →
~/projects/my-app
$ scd scan --deep
Scanning 847 files across 12 modules...
✖ CRITICAL SQL-001 Unsanitised input in user query src/api/users.js:142
⚠ HIGH JWT-003 Weak signing algorithm detected src/auth/tokens.js:38
⚠ HIGH XSS-002 Unescaped output in template src/views/profile.ejs:91
· INFO LOG-001 Sensitive value in log statement src/lib/debug.js:17
✔ Deep analysis complete — no code left your network
Report: ~/.scd/repos/my-app/reports/scan-a3f7b2c1.html

AI writes code
faster than you can review it

Claude Code, GitHub Copilot, and Cursor help your team ship faster than ever. But the same tools that accelerate delivery also accelerate the introduction of security vulnerabilities.

Most SMB development teams have no dedicated security engineer. Security reviews happen — if at all — as an afterthought. By the time a vulnerability reaches production, the cost of finding it has multiplied.

scd puts security in the workflow, not in a separate audit cycle.

OWASP A03
Injection vulnerabilities remain the most common finding — even in AI-generated code.
189+
Rules covering OWASP Top 10, NIS2 and CRA patterns across 5 languages — plus file-format scanning for secrets, config leaks and sensitive data.
0
Lines of code sent outside your infrastructure. Your code stays where you put it.
Git
Hooks on pre-commit and pre-push mean findings surface before they ever reach your repo.

Three layers, one workflow

scd works where development happens — in the terminal, in git, and in your team's daily rhythm. No SaaS subscription, no upload portal.

⌨️

scd CLI

Open-source command-line scanner. Run scd scan in any repository. Git hooks scan automatically on commit and push. Reports in HTML, Markdown, and JSON — stored locally.

Free / Open Source Learn more →
📊

scd-server

Self-hosted team dashboard. Aggregates findings across all repositories and developers. Trend analysis, knowledge gap identification, exception approval workflow, and compliance reporting (CRA, NIS2). Runs inside your own infrastructure.

Team Learn more →
🧠

Deep Analysis

AI-powered analysis of each finding: real issue or false positive? Attack scenario? Exact fix? Available today via Claude API — only the triggering code line is sent, never whole files.

Coming soon — Local AI analysis via a local AI provider: your code never leaves your environment.

Included in Team tier

Your infrastructure,
your control

👨‍💻

Developer

Runs scd scan locally. Findings stay in ~/.scd/

🔍

scd scan

189+ rules applied. Pre-commit hooks intercept secrets. Results stored locally.

🖥️

scd-server

Your server, your network. Aggregates team data. Runs on a VM inside your infrastructure.

🤖

AI Analysis

Deep analysis via Claude API — available today, controlled per-repo by trust level.

Coming soon — Fully offline AI analysis via a local AI provider.

code_left_environment: false  —  auditable in every scan result
  • 🔴
    Injection vulnerabilitiesSQL injection, command injection, LDAP injection — OWASP A03
  • 🟠
    Authentication & session flawsWeak JWT algorithms, missing expiry, insecure session handling
  • 🟠
    Cross-site scripting (XSS)Reflected, stored, and DOM-based XSS in templates and output
  • 🟡
    Secrets & sensitive data exposureAPI keys, passwords, tokens in code and logs — caught at pre-commit
  • 🟡
    Insecure dependenciesKnown CVEs matched against OSV and CISA KEV — locally, privately
  • 🔵
    Infrastructure misconfigurationsHardcoded hosts, unsafe defaults, missing security headers

Built on OWASP
and EU regulations

All rules map to OWASP Top 10, EU Cyber Resilience Act (CRA), and NIS2 directive requirements. Premium tiers generate ready-to-use compliance reports for documentation and conformity assessments.

Rules cover JavaScript, TypeScript, Python, PHP, and ASP.NET — the languages AI coding tools use most. Scanning is context-aware: test files are classified separately so they never inflate production findings.

A01 Broken Access Control A02 Cryptographic Failures A03 Injection A04 Insecure Design A05 Security Misconfiguration A07 Auth Failures A08 Software Integrity A09 Logging Failures CRA Article 13 NIS2 Article 21
Beyond source code

scd also scans configuration and data files for leaked secrets and sensitive data.

.env .json .yaml .toml .log .txt .csv .sql .conf .xml .pem .sh

"We are so serious about privacy that we do not even want your data."

— Core principle, Activemind Solutions AB

🏛️

Self-hosted by design

scd-server runs in your own infrastructure. Scan results, findings, audit logs, and developer activity never leave your network. Activemind has no access to any of it.

🔒

No tracking, ever

No analytics SDKs. No behavioural telemetry. No usage data collection. The only external communication is license heartbeat and binary integrity verification — both documented and auditable.

📋

Full transparency

Every external network call is documented in an integrity manifest. You can verify what leaves your network, when, and why. AI analysis records code_left_environment per finding.

Every scd-server release ships with a signed integrity manifest listing every outbound connection, its purpose, and exactly what data it carries. Network-fence your installation and verify it yourself.

Start free, scale with your team

scd CLI is free and open source — forever. For professional support and team features, contact us for tailored pricing.

Starter
Free
Open Source · Forever
  • scd CLI with 189+ rules
  • Local HTML, Markdown & JSON reports
  • Git hooks (pre-commit + pre-push)
  • Unlimited repositories
  • scd insights (local)
  • AI analysis per finding
  • scd-server
  • Team dashboard
  • Trend analysis
  • Compliance reports
Get started free →
Premium support for CLI
Contact us
Pricing on request
  • Everything in Starter
  • Professional email support
  • Guaranteed response time
  • Direct Activemind contact
  • Security guidance & consulting
  • Onboarding session
  • Team dashboard
  • scd-server
  • AI analysis per finding
Contact us →

More from Activemind

Beyond the product tiers, we work directly with teams that need something extra. Reach out if any of the below sounds relevant — nothing is off the shelf yet, but we are happy to talk.

🎓
Security training Workshops and e-learning tied to your team's actual OWASP knowledge gaps — not a generic course.
🔎
Security review A structured review of your codebase or architecture by Activemind, with a written findings report.
🤝
Ongoing consulting Dedicated Activemind contact, SLA, and regular security check-ins for teams that want a partner.
📦
Custom integrations Tailored scd-server setup, CI/CD pipelines, or compliance evidence packages for your specific stack.

scd CLI is free
and open source

The core scanner, all 189+ OWASP rules, git hooks, and local reporting are free — forever. No account required, no data sent anywhere.

Open source is not a loss-leader or a demo mode. It is a genuine, complete security tool that any developer can run today. The commercial tier adds team infrastructure and professional support.

scd is licensed under AGPL-3.0. Using scd to scan your code does not affect your code's license in any way — the AGPL applies only if you incorporate scd's source code into your own product or distribute it as part of a service. Running scd scan imposes no obligations on your software whatsoever.

View on GitHub
{scd}
activemindsolutions/scd Public · AGPL-3.0
189+ rules 5 languages AGPL-3.0
$ npm install -g @activemind/scd
✓ Node 22+

Ready to secure
your codebase?

scd is developed and supported by Activemind Solutions AB — a Swedish security consultancy specialising in penetration testing, security audits, and developer training.

Pilot pricing available for the first five customers. Ask about founder pricing.