TPB HOME

Security intelligence / project 02

Alfred

Live evidence.
Practical risk.

Custom collectors, firewall APIs and AI-assisted interpretation turn security logs into a clearer picture of what actually happened.

Active DirectoryFirewall + SYSLOGIOC analysis
Conceptual illustration of directory, firewall and log evidence passing through an analysis lens into a clear report.

Collect · correlate · explain

The challenge

An alert describes a signal. The surrounding evidence explains its significance.

A generic indicator description can explain why something is suspicious. Alfred asks what happened in this event: was the activity blocked, did communication continue, was there meaningful recurrence, and what can the evidence actually establish?

System design / two complementary paths

Continuous visibility.
Focused incident analysis.

Monitoring & daily reporting

AD + firewall + SYSLOG

Custom collectors bring Active Directory and security-log evidence into a unified reporting workflow. This provides a broader operational view across identity and network activity.

Event-driven IOC analysis

Trigger → evidence → assessment

An indicator of compromise (IOC) starts a dedicated workflow. API retrieval collects relevant firewall/SYSLOG evidence around the event, followed by classification, contextual enrichment and AI-assisted reporting.

The IOC path is separate from daily reporting. Near-real-time means analysis follows an incoming event; it is not a claim of instantaneous streaming or a guaranteed response time.

  1. Receive

    Capture the IOC trigger and its available event context.

  2. Retrieve

    Query current firewall/SYSLOG evidence through custom API-backed collectors.

  3. Assess

    Apply deterministic risk rules and correlate endpoint evidence where attribution is supported.

  4. Interpret

    AI explains the evidence, practical significance and remaining uncertainty.

  5. Report

    Deliver a structured incident report for human review.

Logical workflow, simplified for this case study. Collection services make the API requests; the AI reporting layer consumes their evidence. No production topology or endpoint details are shown.

From log severity to operational significance

What does this event mean
in the real world?

The source alert and Alfred’s practical assessment stay distinct. A concerning indicator may describe an attempted connection that was contained, or one followed by successful traffic and stronger evidence of compromise.

Lower practical concern

Contained activity

  • The triggering activity was blocked.
  • No successful follow-on is observed.
  • No stronger attack or malware evidence is present in the collected window.

A lower risk assessment still describes the limits of the available evidence. It does not prove the endpoint is clean.

Elevated practical concern

Behavior worth investigating

  • Meaningful activity recurs in separate episodes.
  • Related activity appears across IOC history.
  • Successful post-trigger communication warrants review.

The explanation shows what changed the priority and why a human should investigate.

Higher practical concern

Stronger compromise evidence

  • Attack or malware evidence is present.
  • Serious security activity was not contained.
  • Malicious activity is supported by successful follow-on.

The assessment identifies the evidence supporting escalation and recommends human review.

Illustrative risk factors, not an exhaustive rule table. Actual classification depends on the evidence combination. Source severity is preserved; deterministic rules control Alfred’s practical tier.

The role of AI

Explain the evidence.
Make the judgment readable.

AI turns the collected facts and computed risk into an event-specific narrative: the best-supported trigger, why it matters, what remains uncertain and what to review next.

Grounded in the current event

The explanation uses the collected firewall/SYSLOG evidence and available device context. Generic web-based descriptions of an IOC are background information; they cannot establish what happened in this environment.

Rules control the risk tier

The language model is explanatory. It cannot override the deterministic classification or promote a confident-sounding narrative above the underlying evidence. This separates repeatable risk decisions from natural-language interpretation.

Uncertainty stays visible

A likely trigger is not automatically proof of an internal detection signature. A hostname is not proof of a person’s identity, and a current device association is not proof of its exact location when the event occurred.

What I built

The engineering behind
the interpretation.

The work connects source-specific collection, event processing, deterministic evidence handling and local AI reporting into a usable security-monitoring system.

Custom collectors

Built source-specific collection workers for directory/security monitoring and IOC intake, keeping collection logic separate from the AI prompt and reporting layer.

API integration

Used the event context from each IOC trigger for targeted firewall/SYSLOG API queries. Retrieved records become structured evidence bundles for classification and reporting.

Evidence engineering

Passed structured JSON results between workers, separating source findings, practical risk and attribution from the generated explanation.

AI-assisted reporting

Used a local language model to turn structured evidence and rule-based risk into a readable assessment with explicit uncertainty.

Episodes, not row counts

Repeated rows from one session are not automatically repeated behavior. Grouping trigger activity into episodes helps prevent a single burst of log noise from inflating practical risk.

Attribution with evidence

Device correlation proceeds only when the identity evidence supports it. Ambiguous endpoint information remains unproven rather than becoming a confident identity claim.

Preserve the evidence

The reporter consumes collection and classification outputs. Raw findings remain authoritative and can be checked against the explanation.

Build & validation

Improved against observed behavior.

Trace the wrong priority

Investigated risk inflation caused by unrelated log activity and by treating repeated records as separate threat episodes.

Repair the evidence chain

Refined endpoint attribution and reporting sequence so the report can use supported device evidence while preserving uncertainty.

Replay, then validate

Used targeted replays and a fresh end-to-end event to check the corrected behavior before recording a stable release.

The result

A report that explains
why it matters.

The finished output brings together the source finding, best-supported trigger, practical risk, device context, timeline and recommended next steps in a readable PDF delivered to reviewers.

That gives the operator a reasoned account of the event, with the evidence and limitations visible.

Report structure / illustrative

Source finding

What the security system reported.

Practical risk

The rule-based assessment and its supporting facts.

AI explanation

What the event means, what is uncertain and what deserves attention.

Next review

Evidence to inspect and human decisions to consider.

Evidence-led interpretationHuman-reviewed responseNo automated containment