SecureMeshRequest a demo

Defensive monitoring · read-only · analyst-reviewed

Security alerts that deserve attention.

SecureMesh is a proposed platform that consolidates authorised security findings, correlates them into incidents and drafts evidence-based summaries — so analysts spend their time on what matters, and decide every response.

Findings in
16
Incidents out
8
Need attention now
2

Sample day from the prototype, synthetic data.

mesh · 12 assets · 24h

// from noise to signal

Sixteen alerts. Eight incidents. Two that can’t wait.

Findings from different tools describe the same event. SecureMesh groups them by shared resource, identity and time, removes duplicates and ranks what is left — each step shown, nothing hidden.

stream · unsorted · as received
  1. GuardDutyConsole login from an unusual locationiam-user/deploy-bot
  2. GuardDutyAPI calls from an unusual networkiam-user/deploy-bot
  3. Security HubBucket policy allows public reads3/customer-exports
  4. CloudWatchSpike in S3 GetObject requestss3/customer-exports
  5. GuardDutyPort probe on an exposed portstaging-web-01
  6. GuardDutyPort probe on an exposed portstaging-web-01
  7. Security HubSecurity group allows SSH from anywherestaging-web-01
  8. GuardDutyDNS requests to a crypto-mining domainbuild-runner-07
  9. CloudWatchCPU at 100% for 40 minutesbuild-runner-07
  10. Security HubRoot account usedroot-account
  11. Security HubMFA not enabled for rootroot-account
  12. Security HubBucket policy allows public reads3/marketing-assets
  13. GuardDutyUnusual database login failuresprod-db-cluster
  14. GuardDutyUnusual database login failuresprod-db-cluster
  15. Security HubTLS certificate expires in 7 daysprod-payments-api
  16. GuardDutyPort probe on an exposed portstaging-web-01

// capabilities

Everything an analyst needs to triage, nothing that acts without them.

0x01

Security event aggregation

Read-only ingestion of authorised findings from GuardDuty, Security Hub and CloudWatch into one normalised stream.

0x02

Defensive anomaly detection

Flags behaviour that departs from each identity’s and asset’s baseline — new networks, sudden request spikes, unusual logins.

0x03

Alert classification & prioritisation

Correlates related findings and ranks incidents P1–P4 from severity, asset criticality, corroboration and confidence.

0x04

Evidence-based incident summaries

A plain-language timeline built only from the findings, with sources — drafted for an analyst to review, never sent on its own.

0x05

Security monitoring dashboards

Open incidents by priority, source and asset; time to acknowledge; what was marked benign and why.

0x06

Incident case management

Analysts open cases, add notes, move them through investigating, contained and closed, with a closing summary.

0x07

Audit reporting & access controls

Analyst, lead, CISO and auditor roles; every decision logged with who and why; exportable reports.

core

The differentiator

AI-assisted alert triage that combines event context with analyst-reviewed incident summaries. The machine does the gathering; people do the judging.

// transparent priority

A score you can argue with.

No black box. Every incident shows exactly why it ranks where it does. Move the inputs and watch the priority change — it’s the same formula the triage console uses.

score = severity/10 × criticality/3
      × (1 + 0.25 × (signals − 1))
      × confidence
priority calculator
P1
84

P1 ≥ 60 · P2 ≥ 35 · P3 ≥ 15 · otherwise P4. Thresholds are configurable per team.

Defensive by design.

  • Read-onlySecureMesh reads findings from sources you authorise. It never changes your infrastructure.
  • No automated responseNo blocking, quarantining or account changes. Analysts decide and act in their own tools.
  • Drafts, not verdictsSummaries are built from the findings and labelled as drafts until an analyst reviews them.
  • Every decision explainedMarking an incident benign or closing a case needs a written reason, kept for audit.

// built for

Teams who watch production for a living.

IT

Enterprise IT departments

One queue across accounts and tools, with priorities leadership can follow.

SaaS

SaaS companies

Small security teams covering fast-changing cloud estates without drowning in alerts.

MSSP

Managed security providers

Many customer environments, consistent triage and audit-ready reporting for each.

Decision-makers CISOs · security analysts · IT infrastructure managers

// business model

Priced by environments and data volume.

Subscriptions based on the number of monitored environments, with 5 GB/day of findings and logs included per environment and usage beyond that billed by volume. Illustrative pricing for a proposed product.

estimate

$750/ month · $720 for environments + $30 for 5 GB/day over the included volume

Request a demo

// planned AWS infrastructure

How we use Amazon Web Services.

SecureMesh is a proposed defensive monitoring platform for enterprise IT teams, SaaS companies and managed security providers. It works on a continuous stream of findings from authorised sources, which must be consolidated, explained and kept as protected records. Designing on AWS lets it read findings where they already live, process them as they arrive and store them with strict access controls.

  1. 01

    Findings

    Amazon GuardDutyAWS Security Hub

    Threat detection findings and aggregated security findings are read from the customer’s authorised accounts, read-only.

  2. 02

    Operational signals

    Amazon CloudWatch

    Operational alarms and logs add context, such as a deployment or an outage at the same time.

  3. 03

    Process & correlate

    AWS Lambda

    Findings are normalised, de-duplicated and correlated into incidents by resource, identity and time.

  4. 04

    Summarise

    Amazon Bedrock

    Evidence summaries are drafted from the correlated findings for analysts to review.

  5. 05

    Protected records

    Amazon S3

    Findings, incidents, case notes and audit reports are stored encrypted with restricted access.

// across every stage

  • AWS Lambda runs ingestion and correlation
  • AWS KMS encrypts security records
  • Amazon Cognito separates analyst, lead and auditor roles
  • AWS CloudTrail records access to protected data

Services and their role in SecureMesh

Amazon GuardDuty
Supplies threat-detection findings from the customer's authorised AWS accounts.
AWS Security Hub
Aggregates security findings and control results across accounts.
Amazon Bedrock
Drafts analyst-facing evidence summaries from correlated findings.
Amazon CloudWatch
Provides operational alarms and logs as context for incidents.
Amazon S3
Stores findings, incidents, case notes and audit reports as protected records.
AWS Lambda
Runs alert ingestion, normalisation, de-duplication and correlation.
Amazon Cognito
Authenticates users and enforces role-based access.
AWS KMS
Encrypts security records with customer-specific keys.
AWS CloudTrail
Keeps an audit trail of access to protected resources.
AWS Activate · company description

SecureMesh is a proposed defensive cybersecurity monitoring platform for enterprise technology teams, SaaS businesses and managed security service providers. The planned application will consolidate authorised security findings, prioritise alerts and generate evidence-based investigation summaries for analyst review. We plan to use Amazon GuardDuty, AWS Security Hub and Amazon CloudWatch as sources of security and operational findings, alongside Amazon Bedrock for summarisation. Amazon S3 and AWS Lambda will support protected record storage and processing. Security professionals will evaluate findings and determine incident response actions.

Planned, not yet deployed. This website and the triage console run in your browser on synthetic data.

// founder

David Emeka

David founded SecureMesh after seeing security teams lose real incidents in a flood of duplicate, unexplained alerts. David leads product, customers and every pilot, with a focus on read-only, analyst-first tooling.

  • Product
  • Alert triage
  • Customer pilots

// progress & demos

See your alert queue get quieter.

SecureMesh is a proposed product in its building stage. A demo walks through the review workflow: findings in, incidents ranked, a summary drafted and an analyst deciding. The MVP covers read-only ingestion of authorised findings, alert categorisation, evidence summaries and manually managed incident cases.

  1. 2026
    Company founded

    By David Emeka, to give security teams fewer, better-explained alerts.

  2. Done
    Triage method defined

    Read-only ingestion, correlation into incidents, transparent priority scores, analyst-reviewed summaries.

  3. Done
    Triage console prototype

    Findings, incidents, evidence summaries, case management and audit export on synthetic data.

  4. Done
    AWS reference architecture

    Finding sources, aggregation, summarisation, storage and processing designed.

We use these details only to respond to your request. See the privacy notice.

// faq

Questions security teams ask.

Does SecureMesh respond to threats automatically?

No. It is read-only and never blocks, quarantines or changes anything. Analysts review incidents and decide every response in their own tools.

What data does it read?

Only findings and logs from sources you explicitly authorise, such as GuardDuty, Security Hub and CloudWatch in your own AWS accounts.

Are the summaries written by AI?

In the planned product, Amazon Bedrock would draft summaries strictly from the correlated findings. They are labelled as drafts until an analyst reviews them. The prototype uses a deterministic template.

Is SecureMesh live?

Not yet. It is a proposed product; the triage console is a prototype running in your browser on synthetic findings.

// contact

Talk to SecureMesh.

Questions about the product, integrations, data handling or partnerships. Messages go directly to the founder.

david@securemesh.space