The LESIS Platform

One AI-native cybersecurity platform for your entire program.

Vulnerabilities, code, attack surface, knowledge and incidents.
AI verifies every finding before it reaches your team, and LESIS engineers stand behind every decision.

The LESIS security posture page: findings past their deadline, time to fix and the attention points that need a decision.

Organizations we've helped protect

  • Adobe
  • UOL
  • Oracle
  • Ford
  • D-Link
  • TIM
  • OLX
  • Smart Fit
  • Rentcars

The platform

Everything your security program runs on.

Every capability runs on the same data, so a finding in code, a threat model and an incident all point to the same systems and the same people.

  1. 01 · Posture and analytics

    Know where you stand at any moment.

    Twelve months of findings opened, fixed and closed, time to fix by severity, and the attention points that need a decision now, each with why it matters and what to do next.

    • Security posture and remediation health
    • Attention points with a next step
    • The value of what your program found
    • Benchmark against other organizations on LESIS
  2. 02 · Vulnerability management

    Every finding, from first evidence to proof of fix.

    Evidence, comments and the attack tree live on the finding, and each one has an owner and a deadline set by its severity. Remediators move it through eight states; only an analyst can close it.

    • Deadlines per severity, set by your organization
    • Attack trees on the finding
    • Share one finding with an outside guest
    • Automatic retests for code findings
  3. 03 · Code security

    Code review that has to prove it.

    Connect GitHub, GitLab or Bitbucket with read-only access. AI maps each application, follows the data to where it is used and reports only what a second, stronger review confirms. Secrets are judged without the AI ever seeing the value.

    • Read-only connections, kept in sync
    • AI code review, verified before it reports
    • Secret scanning that keeps the value from AI
    • Only what changed is read again
  4. 04 · Exploit chains and root causes

    See how findings chain into real attacks.

    AI correlates open findings into attack paths and groups them by the correction they share. Analysts edit, add or dismiss each step, and every change is marked with who made it.

    • Attack paths across findings and assets
    • Root causes that close several findings at once
    • Steps edited or added by people, marked as such
    • Full-screen present mode and a YAML editor
  5. 05 · Attack surface

    Every asset, repository and how they connect.

    An inventory of what you own and where the code lives, drawn as a graph with the open findings on each resource. Internet-facing systems are marked, and links you record are kept apart from links the analysis derives.

    • Assets and repositories, linked
    • Open findings on each resource
    • Internet-facing systems marked
    • Recorded and derived relationships kept apart
  6. 06 · Knowledge

    Your policies and threat models, linked to the work.

    Policies, procedures, playbooks, runbooks and threat models live next to the findings they explain. Mention a finding, asset, repository or investigation and it is linked to the document. Attack trees are drawn right in the document.

    • Policies, procedures, playbooks, runbooks and threat models
    • Mentions that link findings, assets and repositories
    • Attack trees drawn in the document
    • Search across every document
  7. 07 · Your advisor

    A named engineer, one click away.

    Your organization gets a LESIS advisor inside the platform: a security engineer who knows your systems, reviews what the AI proposes and joins your team when a fix needs a second pair of eyes. Email them or book a call without leaving LESIS.

    • A named LESIS engineer for your organization
    • Profile, email and booking link in the app
    • Reviews AI proposals and fixes with your team
    • On the home screen, beside the AI assistant

AI with guardrails

It reads everything.
It reports only what it can prove.

Most AI security tools hand you every suspicion they have. In LESIS, each one goes through a second, stronger review that has to show how an attacker gets in, and an analyst confirms what reaches your team.

  1. It never sees a secret

    Candidate credentials are judged from the code around them. The value is masked on screen and removed before code reaches a model.

  2. It can't change your code

    LESIS connects to your repositories read-only, so no model and no one at LESIS can push a commit, open a pull request or touch a branch.

  3. AI proposes, people decide

    Exploit chains, root causes and suggested records stay drafts until an analyst confirms, edits or dismisses them, and every change shows who made it.

  4. Proof of fix, too

    When a code finding moves to Retest, AI checks it against the current code and posts the result on the finding.

API

Your program, programmable.

A documented REST API covers the platform, so findings flow into your tickets, your SIEM and your data warehouse, and your own AI agents work with the same records your team sees, under the same permissions.

  • A documented REST API across the platform
  • Keys bound to one organization and their owner's permissions
  • Scoped permissions, expiry dates and revocation
  • A guide for AI agents, downloadable as Markdown

Coming soon

  • Soon

    LESIS CLI

    Query and update your program from the terminal and from your CI pipelines.

  • Soon

    LESIS MCP server

    Give your AI assistants and coding agents governed access to your findings, within the permissions of the key.

$ curl -G https://platform.lesis.lat/api/reports \
    -d severity=critical,high -d sla_exceeded=1 \
    -H "Authorization: Bearer $LESIS_API_KEY"

{
  "success": 1,
  "reports": [
    {
      "title": "Account takeover through password reset token reuse",
      "severity": "critical",
      "status": "fixing",
      "assignee_name": "Camila Souza",
      "sla_breached": 1
    },
    …
  ],
  "pagination": { "page": 1, "per_page": 20, "total": 3, "total_pages": 1 }
}

Enterprise

Built for organizations
with more than one of everything.

Subsidiaries, business units, outside researchers and auditors, each with exactly the access they need, and a record of what they changed.

  1. Multi-organization

    Run each business unit or subsidiary as its own organization, with its own deadlines, and see them all together when you need to.

  2. Roles that match how you work

    Eight roles, from administrator to analyst, remediator and guest. Each one is enforced on the server, not just hidden in the interface.

  3. An audit trail

    Changes across the platform are logged with the account that made each one and when.

  4. Guest access

    Share a single finding or investigation with someone outside your organization. They see that record and nothing else.

  5. Benchmark

    Your remediation metrics beside the median and quartiles of other organizations on LESIS, none of them named.

  6. The value of the program

    What your findings would be worth, priced per severity from bug bounty data or from values your organization sets.

  7. Two-step sign-in

    A password and a one-time code sent by email. New accounts wait for an administrator's approval.

  8. Least privilege on your code

    Read-only repository access, GitHub tokens limited to the repository being scanned, and stored credentials encrypted.

  9. Deadlines you set

    Remediation windows per severity for each organization, and an email digest of everything past due.

Experts on call

Engineers behind every finding.

LESIS is built and backed by a security research lab. Your organization has a named LESIS advisor, and the same engineers review what the AI proposes, triage what outside researchers send and run deeper assessments when a system needs one.

  1. Your advisor

    A named LESIS engineer for your organization, reachable from inside the platform, who knows your context.

  2. Decisions AI cannot make

    Analysts confirm or dismiss what the AI proposes, and test what code review cannot reach.

  3. Disclosure triage

    On the managed program, LESIS triages reports from outside researchers as they arrive.

  4. Assessments

    Vulnerability research, architecture review and incident investigations, with every finding landing in the same record.

Ways to run LESIS

Run the platform with your own team, or have LESIS run your program on it with you. Either way the record is yours, and the engineers behind it are the same.

Platform

Your team runs the program on LESIS.

Every capability for your organization: posture and analytics, vulnerability management, code security, exploit chains and root causes, your attack surface and knowledge. AI does the reading, and your LESIS advisor is there when a decision needs a second opinion.

Book a demo

Managed

LESIS runs your security program on the platform, alongside your team.

Findings triaged, chased, retested and closed. Disclosure reports triaged as they arrive. Inventory, runbooks and policies kept current; posture benchmarked against other organizations. You don't add headcount: your engineers step in when a fix needs them, and what they learn goes into the record.

Start a conversation

Consultancies can deliver LESIS under their own brand, or refer clients to us, through the LESIS Partnership Program, without building an internal team.

One platform for the whole program. See it on yours.

In a demo we walk through the platform with a fictional organization, then talk about yours: what you run, who looks after it and where your findings live today.