NewEvery agent tool call now routes through the governed MCP gateway — policy check, audit log, then dispatch.Read the changelogarrow_forward
XOOZLYAI Orchestrator
routeHow it workssmart_toyAgent fleetshieldGovernance & trustsellPricingmenu_bookDocsarticleBloghistoryChangelog
Log inStart free
grid_viewProductsrouteHow it workssmart_toyAgent fleetshieldGovernance & trustsellPricingmenu_bookDocsarticleBloghistoryChangelogHelp centreLog in
Start free
  1. XOOZLY
  2. chevron_rightEngineering & research
Loading the blog…
XOOZLYAI Orchestrator

Ready to orchestrate your AI fleet with total confidence?

Start a project, set a cost cap, and watch the loop run with every tool call policy-checked and logged before it leaves the platform.

Start freearrow_forwardSee what shipped
  • check_circleNo credit card required
  • check_circleYour keys, your model spend
  • check_circleEvery tool call audited

Platform

  • All productsarrow_forward
  • Orchestration looparrow_forward
  • Agent fleetarrow_forward
  • Governed control planearrow_forward
  • Pricingarrow_forward

Trust & legal

  • Trust centrearrow_forward
  • Security & compliancearrow_forward
  • AI use policyarrow_forward
  • Report a vulnerabilityarrow_forward
  • Data subject requestsarrow_forward

Resources

  • Documentation & guidesarrow_forward
  • Engineering & research blogarrow_forward
  • Changelogarrow_forward
  • Uptime & reliabilityarrow_forward

Browse by domain

  • Strategy & Researcharrow_forward
  • Developmentarrow_forward
  • QAarrow_forward
  • Governance & Orchestrationarrow_forward
  • All 10 domainsarrow_forward
XOOZLY
© 2026 XOOZLY · 61 agents · 10 domains · 8 SDLC stationsAI use policyarrow_forwardLog inarrow_forward99.9% uptime SLAarrow_forward
  1. XOOZLY
  2. chevron_rightEngineering & research

12 dispatches·Latest Aug 10, 2026

terminalEngineering & research blog

Insights from the frontier of governed AI

Technical deep-dives on governed agent orchestration: the MCP gateway choke point, deterministic guardrails, hard cost caps, tenant isolation and the eight-station loop — written by the people building them.

Filter dispatches by topic
search⌘K

sortShowing 6 of 12 · newest first

Featured dispatchGovernanceAug 10, 2026 · 8 min read

Why we built a central MCP gateway instead of trusting agents directly

Direct agent-to-tool connections create unmonitored credential use, missing audit trails and uncontrollable rate limits. Here is how one centralized proxy enforces policy, writes the audit row and dispatches — in that order — before a request reaches a third-party server.

PAPlatform Architecture TeamCore protocol runtime & authRead full analysis: Why we built a central MCP gateway instead of trusting agents directlyarrow_forward
Gateway enclave flowGRANTED CALLS ONLY
Agent nodes reach external APIs only through the gateway, which applies a policy check, writes an audit row and then dispatches. The direct path from an agent to an external API is blocked.Agent nodesUngranted fleetMCP GATEWAY1. Policy check2. Audit log3. Connector dispatchGranted call onlyExternal APIsThird-party toolsDirect bypass blocked
Checks before dispatch: 2Audited outcomes: allow · deny · error
  • AI OpsAug 03, 2026

    Inside the Rule Optimizer: proposals a human still has to approve

    How detected failure patterns in autonomous runs become proposed rule edits a human approves — and why approval writes a new version instead of editing one in place.

    Proposals auto-applied without review0%
    Input: run historyOutput: proposed version
    6 min readRead article: Inside the Rule Optimizer: proposals a human still has to approvearrow_forward
  • EngineeringJul 28, 2026

    Deterministic guardrails vs. probabilistic LLM evaluators

    Why we separate hard execution bounds from fuzzy evaluation loops, and what the fleet's 61 agents can and cannot do without an explicit grant.

    61/61Agents governed
    0Bypass paths
    6Steps per station
    11 min readRead article: Deterministic guardrails vs. probabilistic LLM evaluatorsarrow_forward
  • FinOpsJul 19, 2026

    Eliminating runaway inference costs with hard project-level caps

    Inside the per-project spending limit that is read before a run starts rather than reconciled after it finishes, and the auto-pause that follows a breach.

    Worked example — one project's monthAuto-pause threshold 80%
    Spent to date · 78%Hard ceiling · 100%
    5 min readRead article: Eliminating runaway inference costs with hard project-level capsarrow_forward
  • Security & TrustJul 12, 2026

    Tenant isolation as defence in depth, not as a single filter

    Why a shared-schema platform filters every tenant query in application code and then refuses the same rows again at the database — and why the runtime role must not be a superuser.

    shield_lockTWO INDEPENDENT REFUSALSApplication filter + row-level security, on every queryRuntime role is non-superuser, non-BYPASSRLS
    7 min readRead article: Tenant isolation as defence in depth, not as a single filterarrow_forward
  • ProductJul 04, 2026

    From mockup to frontend code: the zero-drift design handoff

    How mapping a mockup's computed values rather than its class names keeps generated designs and shipped frontend code from drifting apart.

    Token sourceOne @theme block
    PrimarySurfaceAccentAlert
    9 min readRead article: From mockup to frontend code: the zero-drift design handoffarrow_forward

6 more dispatches in this stream

Weekly AI safety & orchestration digest

Stay ahead of autonomous system failure modes.

Get deterministic rule designs, agent post-mortems and compliance protocols delivered every Tuesday.

One email every Tuesday, once the digest ships.

Or read the changelogarrow_forward