We codified decades of
operational expertise.
Then we open sourced it.
Evos are the first to translate and codify decades of operational domain expertise into agent-ready skills.
Every industry runs on expertise that took decades to build.
Humans have always acquired and stored the most valuable expertise in their fields. Yet, it’s been uncaptured by technology — locked in the heads of the people who do the work.
196 years of expertise — captured
Compatible with
Five kinds of capability.
What a capability is
- 01
Skills
The know-how to do a specific task the way an expert would — read a bill of lading, grade a return, classify a tariff line.
- 02
Expertise
The domain context that makes a skill correct: the rules, regimes, and patterns a seasoned operator carries in their head.
- 03
Decision logic
When to act, when to wait, when to escalate — the judgment calls captured from how your best people actually decide.
- 04
Actions
The things an operator can do in your systems — send, file, book, update, escalate — each one named, scoped, and reversible.
- 05
Workflows
Skills, logic, and actions composed into an end-to-end procedure that runs a piece of work from trigger to resolution.
Benchmarked performance
Every capability is eval-tested against real operational scenarios.
8 sectors across 4 industries, measured over 201 eval scenarios against a Sonnet 4 with no domain capability — and growing.
Capability families
Freight exception management
Works delays, damages, shortages, refused deliveries, and carrier disputes from first signal to resolution. Applies Carmack Amendment filing logic, assesses financial exposure, and routes each case down the right path — write-off, claim, or escalation.
+9.8pp over baseline across 30 scenarios
What it covers
- Carmack Amendment filing
- Financial impact assessment
- Carrier-specific escalation
- Dispute resolution routing
Carrier relationship management
Runs the carrier book — rate negotiation, performance scorecarding, tender acceptance, routing-guide construction, and RFP evaluation across asset carriers, brokers, and niche specialists.
+9.0pp over baseline across 22 scenarios
What it covers
- Rate negotiation logic
- Performance scorecarding
- Routing-guide construction
- RFP evaluation
Customs & trade compliance
Navigates HS tariff classification, free-trade-agreement utilisation, restricted-party screening, and documentation requirements across US, EU, UK, and Middle East customs regimes.
+15.8pp over baseline across 28 scenarios
What it covers
- HS tariff classification
- FTA utilisation
- Restricted-party screening
- Multi-regime documentation
Inventory & demand planning
Runs forecasting, safety-stock calculation, reorder logic, and promotional-lift estimation. Handles seasonal transitions, ABC/XYZ segmentation, and vendor minimum-order constraints.
+8.3pp over baseline across 24 scenarios
What it covers
- Demand forecasting
- Safety-stock calculation
- ABC/XYZ segmentation
- Vendor MOQ constraints
Returns & reverse logistics
Manages return authorisation, inspection grading, and disposition routing — restock, refurbish, liquidate, or destroy — while detecting wardrobing, swap fraud, and serial-returner patterns.
+17.7pp over baseline across 24 scenarios
What it covers
- Disposition routing
- Fraud detection
- Inspection grading
- Return authorisation
Production scheduling
Sequences jobs, optimises changeovers, resolves bottlenecks, and re-sequences on disruption. Applies drum-buffer-rope logic, SMED principles, and labour skill-matrix constraints.
+7.4pp over baseline across 23 scenarios
What it covers
- Job sequencing
- Changeover optimisation
- Drum-buffer-rope logic
- Labour skill matrix
Quality & non-conformance
Investigates non-conformance reports, runs root-cause analysis through 5 Whys and Ishikawa, manages the CAPA lifecycle, and applies statistical process control — control charts, capability indices, and special- vs common-cause distinction.
+8.2pp over baseline across 26 scenarios
What it covers
- NCR investigation
- Root-cause analysis
- CAPA lifecycle
- Statistical process control
Energy procurement
Optimises tariff structures, manages demand charges through load shifting and peak shaving, evaluates renewable power-purchase agreements, and navigates deregulated-market procurement across ISO/RTO regions.
+18.0pp over baseline across 24 scenarios
What it covers
- Tariff optimisation
- Load shifting & peak shaving
- Renewable PPA evaluation
- ISO/RTO market navigation
How it stays trustworthy
Every capability is governed.
A capable operator is not enough — it has to be one you can trust to act. Four guarantees hold for every capability in the library.
- 01
Built from real operators
Not scraped, not generated. Every capability is captured from people who have spent years doing the work, through structured elicitation drawn from cognitive and behavioural science.
- 02
Named, versioned, governed
Each capability is named and versioned, with its own guardrails — cost, rate, and the threshold above which it needs sign-off. You can see exactly what an operator can do, and when it changed.
- 03
Deny-by-default
An operator can only use the capabilities on its manifest. It calls nothing it has not been explicitly granted — no surprise actions, no scope it was never given.
- 04
Monitored, self-tightening
Every capability is monitored in production. When success on one starts to slip, the operator automatically routes that work back to human approval until it recovers.
See it in context
The operators these capabilities power.
Browse the full catalogue of role areas an operator can own, across every legacy industry.