AEGIS
$./aegis init--workflows=modernization,itsm--governance=strict

AEGIS_AI

[ enterprise · ai · governance · platform ]v1.0.0 · solarized · console

Most enterprises don’t fail at AI adoption.
They fail at AI accountability.

Aegis is governance-first AI for enterprise IT — every call audited, every token attributed, every response governed by default. Built around the two highest-leverage workflows: legacy modernization and service-desk automation.

❯open dashboard›case study$hire builder

tip: press ⌘K anywhere to navigate, search, generate api keys, run a governance scan, or export the audit log.

─platform · by design
shadow aigoverned

every model call routes through one governed path

legacy code→ python

cobol · fortran · pl/i · rpg · assembly, with docstrings + tests

cost attributionper-token

every call costed at the token level, per workflow

audit trailcomplete

every invocation logged, searchable, and exportable

─workflows · try them live
$ ls workflows/
├──
modernize cobol → python/❯ try

Translate COBOL, FORTRAN, PL/I, RPG, and Assembly into modern Python with AI-generated docstrings and tests. Every transformation is logged.

6+ langs│gpt-4o│≈1.2k tok/job
└──
triage itsm tickets/❯ try

ITIL v4 incident copilot. Every response carries governance metrics, cost attribution, and PII compliance baked in.

itil v4│gpt-4o-mini│token-level cost
─platform · what's inside
$ man aegis
›dashboardreal-time kpis, spend, latency, streaming activity feed›modelsregistry: every governed model with risk tier + 30d usage›policies10 governance rules — toggle persists locally›auditsearchable, filterable event trail with full payloads›integrations20+ connectors — slack, jira, servicenow, datadog, openai›settingsapi keys, team, notifications, billing, danger zone›statuslive health, 90-day uptime per service, incident history›pricingstarter, growth, enterprise — start-trial dialog wired
─governance dashboard · live sample
$ai_spend.csv · last 6 months · USD
$213
Jan
$340
Feb
$477
Mar
$499
Apr
$523
May
$608
Jun
modernizationitsmΣ $2,660 · 6mo
$tail -f /var/log/aegis/audit.loglive
14:23:18[OK]platform-eng: modernize.cobol_to_python · 1,204 tok · $0.012
14:22:51[WARN]compliance: pii.scan flagged ssn-like pattern · auto-redacted
14:22:04[OK]service-desk: itsm.triage · gpt-4o-mini · 312 tok · $0.003
14:18:42[OK]platform-eng: modernize.fortran_to_python · 847 tok · $0.008
14:14:11[OK]service-desk: itsm.escalate · routed P1 · sla-bound 15m
$
─package.json · stack
$cat aegis/package.json

{

"name": "aegis-ai",

"version": "1.0.0",

"dependencies": {

"next.js": "framework",

"vercel ai sdk": "ai orchestration",

"openai gpt-4o": "llm provider",

"supabase": "audit logging",

"tailwind css": "styling",

"typescript": "language",

"recharts": "viz",

"shadcn/ui": "primitives",

"cmdk": "command palette",

"sonner": "toasts"

}

}

─next steps
$case-study.md

$ less CASE_STUDY.md

Problem · approach · architecture · outcomes

The full write-up: where governance broke, what I built, what shipped, and what I’d do differently next time.

❯ read it
$whoami --hire

$ contact aurimas --role=fractional-pm

Hire the builder · fractional ai pm & architect

Available for governance-first AI engagements: legacy modernization, ITSM copilots, and platform architecture.

❯ open dossier