Features

Leonidus covers the full loop: find the problem, explain the fix, prove the posture. Here's what that means in practice.

Scanning

Every scan runs industry-standard SAST rules plus Leonidus's own rule set, tuned against real detection gaps.

API keys, tokens, and credentials found in source and configuration before they leak.

Known CVEs in packages and images, with severity, exploitability, and upgrade paths.

Point Leonidus at a running app: TLS posture, security headers, exposed endpoints, and known vulnerabilities — with domain-ownership checks built in.

Cron-style schedules per repo or URL, with change-only notifications so quiet weeks stay quiet.

Each finding gets a plain-language 'why this is bad' and 'what to do right now', so fixes don't wait on a security specialist.

The same vulnerability keeps the same identity across rescans — history, suppressions, and diffs stay accurate.

Findings carry owners and due dates computed from severity. Breaches surface automatically.

Silence a false positive with a justification and an expiry — never with a shrug.

SOC 2, HIPAA, ISO 27001, PCI DSS, CMMC, FedRAMP, NIST, state privacy laws, and more — each broken into controls with remediation guidance.

Automatable controls are evaluated from real scan evidence, not questionnaires.

Point-in-time posture snapshots reviewed first by AI, then by a human — with gaps called out per control.

The rest of the audit binder lives here too: policy attestation, vendor risk tiers, and periodic access reviews.

One API call returns pass, warn, or block against your policy. Wire it into CI and stop shipping known-bad builds.

Webhook-triggered scans on every pull request, with results posted back as a comment.

Everything the UI does, the API does. The remote MCP server brings scans into Claude Code, Cursor, and any MCP client.

Executive PDF reports, SARIF for your tooling, SBOM exports, and compliance matrices in XLSX.