SECOPS

MFA-Bypass and New-Geography Triage Agent

An agent that reviews Axiom successful-login events for sessions that skipped MFA or originated from a country the user has never used, gathers context, and decides whether…

CategorySecOps
Enginepaperclip
Difficultyadvanced
Triggerevent
Steps6
Setup~25 min

How it runs

The automated pipeline, trigger to output.

  • TriggerAxiom flags MFA-skip or new-geo loginAxiom
  • ActionPull user's historical geo and device baseline
  • LogicAgent classifies risk: benign, step-up, escalate
  • ActionWrite rationale and evidence for escalations
  • ActionOpen Linear issue for high-risk verdictsLinearLinear
  • OutputPost verdict to secops Slack threadSlack

What it does

Triages suspicious successful logins where MFA was bypassed or the geography is brand-new for that identity. Instead of a fixed rule, an agent weighs the user's history, the auth method, and the resource accessed, then chooses an action and explains why.

When to use it

Use when raw geo/MFA rules generate too many false positives and you want judgment in the loop. Best for orgs with a real history baseline per user in Axiom.

How it works

  1. 1Axiom emits successful logins flagged as MFA-skipped or from an unseen country.
  2. 2The agent pulls the user's historical login geographies and device list for baseline context.
  3. 3It reasons over the signal: known travel pattern, corporate VPN egress, or genuine anomaly.
  4. 4A decision branch routes the outcome — benign events are suppressed with a note; medium-risk events trigger a step-up re-auth request; high-risk events escalate.
  5. 5For escalations the agent writes a plain-language rationale and the supporting evidence.
  6. 6It opens a Linear issue with the rationale and posts the verdict to the secops Slack thread.

Set it up

What you configure once, before turning it on.

  1. 1
    Connect AxiomLog streams, queries, dashboards.
  2. 2
    Connect LinearIssues, projects, cycles, triage.
  3. 3
    Connect SlackChannels, DMs, threads, mentions.
  4. 4
    Set each agent's modelWe leave models unset so you pick the tier — fast + cheap, or top-quality.
  5. 5
    Tune it to your dataEdit the prompts, filters, and field mappings so it matches how your team works.
  6. 6
    Test, then turn it onRun once against a sample, confirm the output, then enable the trigger.

Run this workflow in your colony.

14-day trial. No DevOps. No Sales call. Provisioned in under a minute.