aequai ~/services · ai evidence operations book ↗
aequai ~ / services
$ aequai services --describe --boundary

Services that make AI use reviewable.

Move from scattered AI use to governed workflows. Each service hands you a concrete artifact, a map, a gate, a log, or a decision brief, and ends with one honest call: proceed, shrink, park, or kill.

Readiness & diagnostic Boundary-clear scope Reviewable artifacts
// common deliverables: Each scoped service can produce a typed artifact map, operation boundary, provenance/evidence map, breaker tests, gate record, and transition recommendation. The recommendation names the safe next step, not a public approval or production authorization.
// boundary: These services are readiness, diagnostic, workflow-design, governance, simulation, and decision-support sources. They are not external validation, production implementation, external action, regulated execution, or approval by a public authority or platform partner.
$ aequai start-here --decide

Start where your next decision is unclear.

Choose the smallest service that clarifies the next step. The services read as a ladder, not a menu of unrelated products.

01

Can't name the workflow yet?

Start with the AI Adoption Diagnostic, inventory current AI use, risk, evidence needs, and adoption level.

02

Agents already have tool authority?

Start with the Agent Operations Safety Audit before access expands.

03

One department has a clear owner and workflow?

Use Department AI Pilot Design to scope a controlled pilot.

04

AI use spreading without shared rules?

Add the AI Governance Starter Kit for rules, owners, and escalation.

05

Pain is product data, supplier evidence, or carbon fields?

Use the DPP Readiness Review and consider the Simulation Lab.

06

Buyer starts with tokenization or verifiable records?

Use the Tokenization Readiness Review and test no-token alternatives first.

07

Workflow touches external messages, public wording, or production?

Simulate it in the AI Workflow Simulation Lab before real authority.

$ aequai services --compare

The catalogue at a glance.

ServiceStart whenCore outputNot included
AI Adoption DiagnosticAI use is spreading but the safe first workflow is unclearAI inventory, workflow map, adoption level, evidence and approval needsTool rollout, production build, public claim approval
Agent Operations Safety AuditAgents can read, write, submit, deploy, message, or change systemsTyped artifact map, operation boundary, provenance/evidence map, breaker tests, gate record, transition recommendationExploit testing, credential handling, production operations
Department AI Pilot DesignOne department has a clear workflow owner and measurable painPilot boundary, metrics, simulation plan, implementation-partner briefMulti-department transformation or build execution
AI Governance Starter KitAI use is spreading without shared rulesPolicy skeleton, responsibility matrix, risk tiers, escalation planFormal assurance program or counsel-owned conclusion
DPP Readiness ReviewProduct, supplier, carbon, or disclosure evidence needs structureField readiness, supplier gap map, carbon questions, platform briefOfficial passport operation, public green-claim approval, carbon-asset action
Tokenization Readiness ReviewA token, chain, anchoring, or verifiability idea needs a safe screenNo-token alternatives, rights questions, advisor questions, stop pointsIssuer, broker, custodian, exchange, fundraising, liquidity, or asset execution
AI Workflow Simulation LabA workflow should be tested before authority, sends, or claim-adjacent useSynthetic input pack, agent roles, event log, failure modes, pilot boundaryProduction execution, real credentials, live system mutation, external sends

Each service ends with a decision: proceed (safe enough for a controlled next step with named gates), shrink (narrow the scope first), park (timing, owner, evidence, or advisor not ready), or kill (unsafe, unsupported, or out of scope).

$ aequai services --expand

Six services and one cross-cutting lab.

Expand any service for who it is for, what AequAI reviews, what you receive, where AI fits, where humans stay responsible, and the boundary.

fictional sample  Every sample below is fictional output based on a synthetic company and synthetic inputs. It demonstrates AequAI's review structure and decision format. It is not real customer proof, not a legal opinion, not a compliance approval, and not external validation.

[01]AI Adoption DiagnosticMap current AI use, evidence needs, risk, and the safest first workflow.+ open– close

For founders, COOs, department leads, operations leads, and Mittelstand managers seeing informal AI use spread without a clear operating model.

What AequAI reviews

  • + Current AI use by role, workflow, tool category, data class, and output destination
  • + Workflow opportunity, pain, frequency, reversibility, and evidence availability
  • + Data sensitivity, excluded data classes, and approval boundaries
  • + AI literacy and responsibility gaps by role

What you receive

  • + AI use inventory and workflow opportunity map
  • + L0–L4 adoption-level recommendation
  • + Data sensitivity and approval-boundary map
  • + Evidence and logging needs, plus a role-specific literacy question set
  • + Proceed, shrink, park, or kill recommendation with a 30-day next step

Where AI fits / where humans stay responsible

  • + AI summarizes usage, classifies patterns, drafts maps, as a reviewed draft
  • Humans own workflow selection, data boundaries, source acceptance, and the final decision

Not included

  • Course credentials, tool resale, implementation, or production operation
  • External action approval, external validation, or a regulated conclusion

fictional sample  Northline Components: a synthetic manufacturer narrows informal AI use into one L1 sales-account briefing pilot, shrinks engineering summarization to internal assist, and parks supplier scoring until evidence and ownership improve.

map your current AI use

[02]Agent Operations Safety AuditMap what agents can read, draft, change, send, and escalate before authority grows.+ open– close

For AI-heavy operators, CTOs, engineering leads, support-automation owners, and operations teams using agents that can touch meaningful systems.

What AequAI reviews

  • + Agent and tool inventory, system and data surfaces
  • + Action classes: read, summarize, draft, prepare, write, send, deploy, approve, escalate, mutate
  • + Permission scope, logs, rollback, exception handling, and escalation paths
  • + Rejected and parked actions

What you receive

  • + Typed artifact map for agent inputs, outputs, tools, logs, and review records
  • + Operation boundary for what agents may read, draft, prepare, change, send, deploy, approve, or escalate
  • + Provenance/evidence map, human checkpoint map, permission review, and rollback notes
  • + Breaker tests, rejected-use list, failure-mode notes, and remediation backlog
  • + Gate record and transition recommendation by workflow

Not included

  • Exploit testing, security assurance, or credential handling
  • Production operation, autonomous customer messaging, or deployment approval

fictional sample  Atlas Ops Studio: a synthetic software-ops team lets logged ticket summarization and draft remediation proceed, shrinks repo-write access to branch-only drafts, and rejects autonomous customer replies and production-setting changes.

audit agent authority

[03]Department AI Pilot DesignOne department workflow with owners, metrics, gates, and excluded actions.+ open– close

For department leaders in sales, marketing, support, operations, finance, HR, engineering, procurement, sustainability, or leadership who want one practical AI pilot.

What AequAI reviews

  • + Candidate department workflows: pain, frequency, owner clarity, sensitivity, reversibility, value
  • + Inputs, outputs, blocked outputs, owners, reviewers, and approvers
  • + Pilot metrics, event-log needs, and simulation requirement

What you receive

  • + Department workflow selection and explicit non-goals
  • + Pilot boundary with inputs, outputs, owners, and excluded actions
  • + Approval gates, autonomy-level map, and a metric plan
  • + Simulation-first recommendation, event-log design, and a 30-day pilot roadmap
  • + Implementation-partner brief

Not included

  • Multi-department transformation or build execution
  • Production integration, external messaging, or proof of performance

fictional sample  Meridian Field Sales Team: a synthetic B2B sales department runs a four-week meeting-brief pilot while CRM changes stay draft-only and outbound follow-up automation is parked.

design one department pilot

[04]AI Governance Starter KitPractical AI rules, responsibility maps, risk tiers, and escalation paths.+ open– close

For founders, COOs, people/operations leads, and legal- or compliance-adjacent owners whose teams use AI without shared operating rules.

What AequAI reviews

  • + Current AI uses, policies, approval customs, and sensitive-data boundaries
  • + Allowed, gated, and blocked uses by adoption level and risk tier
  • + Role responsibilities for owners, reviewers, approvers, operators, data owners, and advisors
  • + Public-wording, product-data, and official-approval claim risks

What you receive

  • + AI use policy skeleton and adoption-level definitions
  • + Responsibility matrix and approval-gate model by risk tier
  • + Claim-safety and public-language review workflow
  • + Data and privacy boundaries; incident, exception, and escalation plan

Not included

  • A full assurance program or formal legal conclusion
  • Public approval claims or a replacement for qualified advisors

fictional sample  Solen Works: a synthetic industrial-services company proceeds with a minimum policy skeleton, role matrix, and 30-day review cadence while HR decision-support workflows are parked and public copy stays draft-only.

create minimum viable AI rules

[05]DPP Readiness ReviewPrepare product data, supplier evidence, and carbon fields before a platform.+ open– close

For product owners, supply-chain leads, sustainability leads, importers, brands, manufacturers, and exporters preparing product and supplier evidence.

What AequAI reviews

  • + Product identity, batch, component, material, lifecycle, access, and update fields
  • + Supplier evidence sources, dates, reliability, missing fields, and confidentiality
  • + Carbon evidence fields, method, boundary, assumptions, and source-bundle needs
  • + CBAM and embedded-emissions data-dependency questions where relevant

What you receive

  • + DPP field-readiness matrix and supplier evidence gap map
  • + Carbon evidence field checklist and methodology question list
  • + Green-claim risk questions and public-language review notes
  • + Implementation-partner brief for DPP platform, PLM/ERP/LCA, and advisor handoff

Not included

  • Official passport operation or public sustainability approval
  • Carbon-asset action, platform endorsement, or regulated conclusion

fictional sample  VoltEdge Battery Components: a synthetic battery-component supplier shrinks scope to one product line, runs a supplier evidence sprint, and waits on field readiness before an implementation-partner brief.

map DPP and evidence readiness

[06]Tokenization Readiness ReviewScreen no-token alternatives, rights questions, and advisor gates first.+ open– close

For founders, product owners, asset owners, innovation teams, and operators considering tokenization, anchoring, verifiable records, or chain/vendor choices.

What AequAI reviews

  • + Use-case type: identity, record, access, incentive, asset, claim, verifiability, audit trail, governance
  • + Rights, responsibilities, stakeholders, transfer expectations, lifecycle, and dispute paths
  • + No-token and signed-record alternatives, plus stop points and rejected paths

What you receive

  • + Use-case map and rights/stakeholder question map
  • + No-token alternative and signed-record option
  • + Token-fit decision tree with stop points and a rejected-action list
  • + Counsel and advisor question list; chain/vendor category evaluation questions
  • + Proceed, shrink, park, or kill verdict

Not included

  • Creating or selling tokens, fundraising, liquidity, exchanges, or custody
  • Carbon-asset mirroring or any network/chain action

fictional sample  Harbor Maintenance Ledger: a synthetic maintenance network proceeds with no-token signed records, parks access-pass exploration, and rejects carbon-asset mirroring for the scenario.

screen the token idea safely

[+]AI Workflow Simulation Lab · cross-cuttingSimulate roles, events, agent actions, approval gates, and failure modes before real authority.+ open– close

For any service owner, department lead, operator, or product-data team that needs to test an AI-assisted workflow before real authority, external sends, production changes, public wording, or claim-adjacent evidence use.

What AequAI reviews

  • + Workflow goal, owner, non-goals, and stop conditions
  • + Synthetic or sanitized input pack; agent roles, allowed and blocked actions, human reviewers
  • + Event-log fields, approval gates, failure modes, and rollback notes

What you receive

  • + Simulation design and synthetic/sanitized input register
  • + Agent role and authority map, event log, and approval/escalation map
  • + Failure-mode and rollback notes, before/after comparison
  • + Pilot boundary recommendation and a proceed, shrink, park, or kill verdict

Not included

  • Production execution, real credentials, or live system mutation
  • External sends, official approval work, or regulated conclusions

fictional sample  Supplier Evidence Desk: a synthetic product-data workflow models AI-assisted parsing, gap flagging, human checkpoints, and event logs before any public wording, platform upload, external send, or real workflow authority.

simulate before authority

$ aequai map department × autonomy

Where each department starts, and how far autonomy goes.

An illustrative starting framework. Filled cells show a typical first band. Execute-level actions always sit behind a human approval gate.

departmentL0 assistL1 draftL2 execute +approvalL3 supervised
Customer Support
Marketing & Content
Sales & CRM
Operations
Finance & Accounting
People & HR
// note: This is an illustrative framework, not a recommendation for your business. Your real bands come out of the diagnostic, based on data sensitivity, error cost, and existing controls. The dashed line marks the human approval gate before execute-level actions.
$ aequai path --modular | --integrated

Work modularly, or as one evidence-operations path.

// modular entry

Choose the smallest service that clarifies the next decision. Start anywhere, most teams begin with the diagnostic and move along the ladder as the workflow becomes clear.

// integrated path

Combine AI adoption, governance, DPP/carbon evidence readiness, tokenization screening, and workflow simulation when one workflow has both AI-adoption pain and regulated-evidence pressure.

The integrated path is promising, but buyer preference is still a validation question. AequAI supports modular starting points instead of forcing one large bundle.

$ aequai contact --scope

Not sure which service fits?

Most teams begin with the AI Adoption Diagnostic, then follow the model toward governance and evidence readiness. We will help you sequence it.

Do not send credentials, private keys, private client files, payment data, tax identifiers, legal files, customer data, or sensitive supplier material through public channels.