16 capabilities
Core Controlled Beta
Owner-operated measurement, governance, reporting and decision workflows. Provider requests are not included.
Free during Controlled Beta
Waltrump AI OrchestratorExplore WAO capabilities
Explore tools to evaluate results, understand trade-offs, identify gaps and explain your next action. Browsing does not run a provider request.
16 capabilities
Owner-operated measurement, governance, reporting and decision workflows. Provider requests are not included.
Free during Controlled Beta16 capabilities
Includes Core plus build, evaluation and multimodal workflows. Direct execution requires exact project activation and customer-managed your own provider API key.
Admin-approved4 capabilities
Includes Core and Builder plus private, network and deployment pathways subject to explicit security and provisioning review.
Manual enterprise pilotBuilder BYOK means bring your own provider API key. Provider execution requires separate approval and may incur provider charges; a plan alone does not activate it.
Explore by goal
Higher plans inherit lower-plan access. The authenticated organization view is authoritative for your access.
Showing 36 of 36 capabilities.
See how your AI is performing and what needs checking.
Bring available evaluation results into a scoped health review and see where evidence is missing.
What you get: A review of measured findings and testing gaps to help you choose what to investigate next.
Only measured areas are assessed. This supports your review; it is not certification or a guarantee.
When to use it: Use it when you need a concise view before expanding an AI workflow.
How it works: Bring available evaluation results into a scoped health review and see where evidence is missing. A review of measured findings and testing gaps to help you choose what to investigate next.
Illustrative example: A support assistant has evidence for common questions but not refund exceptions; review the gap before deciding what to test.
Access: Core Controlled Beta. Only measured areas are assessed. This supports your review; it is not certification or a guarantee.
Scoped evidence calculation and saved health review. Input: cross-cutting.
Turn a health finding into an accountable next step.
Record an action against a scoped health finding, then review applicable evidence of completion.
What you get: A saved action and evidence-based verification state for follow-up.
An action is verified only with applicable evidence.
When to use it: Use it when a health review identifies an issue someone needs to address.
How it works: Record an action against a scoped health finding, then review applicable evidence of completion. A saved action and evidence-based verification state for follow-up.
Illustrative example: Assign a coverage gap to an owner and check the supporting evidence before marking it verified.
Access: Core Controlled Beta. An action is verified only with applicable evidence.
Scoped action, verification and saved recalculation. Input: cross-cutting.
See which readiness questions still need supporting evidence.
Provide declarations and available evidence for a scoped readiness review.
What you get: A saved readiness view that separates declared answers from supporting evidence.
Declared answers do not replace measured evidence.
When to use it: Use it when a team needs to organize readiness questions before a review.
How it works: Provide declarations and available evidence for a scoped readiness review. A saved readiness view that separates declared answers from supporting evidence.
Illustrative example: A team records its review process and identifies where evidence has not yet been attached.
Access: Core Controlled Beta. Declared answers do not replace measured evidence.
Declarations, evidence and saved readiness review. Input: cross-cutting.
Review safety signals without treating them as guarantees.
Review safety evidence within the scope of your evaluation.
What you get: A view of available safety signals and their limits for human review.
Safety signals are not a guarantee of safe operation.
When to use it: Use it when safety is part of a decision about an AI workflow.
How it works: Review safety evidence within the scope of your evaluation. A view of available safety signals and their limits for human review.
Illustrative example: Review the evidence from a set of safety cases before deciding which cases need more testing.
Access: Core Controlled Beta. Safety signals are not a guarantee of safe operation.
Scoped safety evidence review. Input: cross-cutting.
Explore how content provenance evidence can be presented.
Inspect a clearly labelled local demonstration using test-root C2PA provenance evidence.
What you get: An illustrative provenance view, not a production publisher-trust verdict.
No production publisher trust or independent authenticity verdict; unknown revocation is untrusted.
When to use it: Use the demonstration to understand the proposed evidence presentation.
How it works: Inspect a clearly labelled local demonstration using test-root C2PA provenance evidence. An illustrative provenance view, not a production publisher-trust verdict.
Illustrative example: Inspect a synthetic asset's test-root provenance without treating it as proof of a real publisher.
Access: Core Controlled Beta. No production publisher trust or independent authenticity verdict; unknown revocation is untrusted.
Clearly labelled local test-root C2PA demonstration. Input: cross-cutting.
No customer activation is offered for this path today.
Review controlled features and share scoped feedback.
Review features within your approved customer scope and submit feedback.
What you get: A record of your review and feedback for the WAO team.
Review does not authorize provider execution.
When to use it: Use it when you want to comment on an available controlled feature.
How it works: Review features within your approved customer scope and submit feedback. A record of your review and feedback for the WAO team.
Illustrative example: Review a beta feature and tell the team where its instructions are unclear.
Access: Core Controlled Beta. Review does not authorize provider execution.
Review and feedback within approved customer scope. Input: cross-cutting.
Keep the evidence behind a decision together.
Use scoped evidence to create and save a decision scorecard.
What you get: A saved scorecard for discussing the evidence and decision with your team.
A scorecard is not WAO certification.
When to use it: Use it when a decision needs an accessible evidence summary.
How it works: Use scoped evidence to create and save a decision scorecard. A saved scorecard for discussing the evidence and decision with your team.
Illustrative example: A release team records its criteria and the evidence it considered in a scorecard.
Access: Core Controlled Beta. A scorecard is not WAO certification.
Saved scoped decision scorecards. Input: cross-cutting.
Bring accountable people into an evidence review.
Review scoped AI evidence with the people responsible for the decision.
What you get: A documented review that helps a team discuss findings and unresolved questions.
Review does not replace representative inputs or approval.
When to use it: Use it when a result needs human judgment before a decision.
How it works: Review scoped AI evidence with the people responsible for the decision. A documented review that helps a team discuss findings and unresolved questions.
Illustrative example: A reviewer considers unusual support responses alongside the available evaluation evidence.
Access: Core Controlled Beta. Review does not replace representative inputs or approval.
Accountable review of scoped evidence. Input: cross-cutting.
See how comparable saved runs vary across repetitions.
Compare eligible saved runs that share a comparable evaluation context.
What you get: A stability view of the comparable evidence available for review.
Only comparable evidence is analyzed.
When to use it: Use it when one result may not show variation across repeated runs.
How it works: Compare eligible saved runs that share a comparable evaluation context. A stability view of the comparable evidence available for review.
Illustrative example: Compare saved runs of the same task to see whether the results are consistent.
Access: Core Controlled Beta. Only comparable evidence is analyzed.
Comparison of eligible saved comparable runs. Input: cross-cutting.
Review a prompt and retain its scoped analysis.
Submit a prompt for native scoped analysis and save the resulting review.
What you get: A saved prompt-analysis result you can reopen for discussion.
No provider request or public SDK publication is included.
When to use it: Use it when refining a prompt before another evaluation.
How it works: Submit a prompt for native scoped analysis and save the resulting review. A saved prompt-analysis result you can reopen for discussion.
Illustrative example: Review an instruction for a support task and keep the result for the next revision.
Access: Core Controlled Beta. No provider request or public SDK publication is included.
Scoped native prompt analysis and exact saved result. Input: cross-cutting.
Consider measured quality alongside costs you actually know.
Review measured quality together with available cost evidence for the scoped workload.
What you get: A quality-and-cost view that makes missing cost information visible.
Unavailable provider cost is never inferred as zero.
When to use it: Use it when comparing trade-offs among measured options.
How it works: Review measured quality together with available cost evidence for the scoped workload. A quality-and-cost view that makes missing cost information visible.
Illustrative example: Review two evaluated approaches using their known costs rather than treating an unknown cost as free.
Access: Core Controlled Beta. Unavailable provider cost is never inferred as zero.
Measured quality and known-cost evidence review. Input: cross-cutting.
Review quality factors supported by your existing evidence.
Use scoped evidence to inspect a basic quality projection.
What you get: A view of quality factors supported by the submitted evidence.
The factor taxonomy is not universal measurement.
When to use it: Use it when you need to discuss output quality with your team.
How it works: Use scoped evidence to inspect a basic quality projection. A view of quality factors supported by the submitted evidence.
Illustrative example: Review measured response-quality factors for a set of support answers.
Access: Core Controlled Beta. The factor taxonomy is not universal measurement.
Basic quality projection from scoped evidence. Input: cross-cutting.
Organize release evidence before a human decision.
Review scoped release evidence and decision-support information.
What you get: A release-oriented evidence view for deciding what still needs attention.
It does not approve or deploy a release.
When to use it: Use it before deciding whether an AI change is ready for release.
How it works: Review scoped release evidence and decision-support information. A release-oriented evidence view for deciding what still needs attention.
Illustrative example: A team reviews missing evaluation evidence before approving a prompt change.
Access: Core Controlled Beta. It does not approve or deploy a release.
Scoped release evidence and decision support. Input: cross-cutting.
Turn submitted evidence into a reviewable report analysis.
Provide evidence for a scoped report-intelligence review.
What you get: A report intelligence result bounded by the evidence supplied.
The report remains bounded by its source evidence.
When to use it: Use it when explaining the implications of evidence to a stakeholder.
How it works: Provide evidence for a scoped report-intelligence review. A report intelligence result bounded by the evidence supplied.
Illustrative example: Review a submitted evaluation summary and its limitations before sharing a conclusion.
Access: Core Controlled Beta. The report remains bounded by its source evidence.
Submitted-evidence report intelligence. Input: cross-cutting.
Keep generated reports available for later review.
Save and retrieve generated reports within your authorized scope.
What you get: A saved report and governed retrieval path for later reference.
Reports are evidence summaries, not certifications.
When to use it: Use it when stakeholders need to revisit a prior generated result.
How it works: Save and retrieve generated reports within your authorized scope. A saved report and governed retrieval path for later reference.
Illustrative example: Open a saved technical report after a team discussion rather than generating it again.
Access: Core Controlled Beta. Reports are evidence summaries, not certifications.
Generated report save and governed retrieval. Input: cross-cutting.
Inspect saved observations and the citations behind them.
Review previously saved observations and citations within the approved scope.
What you get: A view of retained observations and their cited sources.
New external observations require separate exact activation.
When to use it: Use it when explaining what supported a previous observation.
How it works: Review previously saved observations and citations within the approved scope. A view of retained observations and their cited sources.
Illustrative example: Review saved citations for an earlier observation before discussing its implications.
Access: Core Controlled Beta. New external observations require separate exact activation.
Review saved scoped observations and citations. Input: cross-cutting.
Evaluate an agent's proposed steps without executing tools.
Submit a proposal-only trace for a scoped evaluation and save the result.
What you get: A saved trace evaluation for reviewing proposed behavior.
No live tool or computer action is executed.
When to use it: Use it when examining an agent's proposed tool use before any action is authorized.
How it works: Submit a proposal-only trace for a scoped evaluation and save the result. A saved trace evaluation for reviewing proposed behavior.
Illustrative example: Review a submitted trace that proposes a tool call without making the call.
Access: Builder Byok Controlled Beta. No live tool or computer action is executed. Provider execution may require activation and your own provider API key; provider charges may apply.
Submitted proposal-only trace evaluation and saved result. Input: text / trace.
Run a bounded comparison after the required activation.
Set up a bounded Direct run for an authorized project and provider, then recover its saved result.
What you get: A saved result for the eligible run, with failures visible.
The complete three-case comparison is unfinished. Failed cases are not counted as complete.
When to use it: Use it when an approved project needs a scoped provider evaluation.
How it works: Set up a bounded Direct run for an authorized project and provider, then recover its saved result. A saved result for the eligible run, with failures visible.
Illustrative example: An activated team runs an eligible case and reopens the saved result for review.
Access: Builder Byok Controlled Beta. The complete three-case comparison is unfinished. Failed cases are not counted as complete. Provider execution may require activation and your own provider API key; provider charges may apply.
Bounded Direct run and exact saved recovery after project/provider activation. Input: text / cross-cutting.
Prepare a focused benchmark before running any provider.
Create a finite Direct-policy setup and save it as a draft.
What you get: A saved preparation draft that can be reviewed before execution is considered.
Preparation does not execute a provider or certify a Benchmark.
When to use it: Use it when turning an evaluation question into a bounded setup.
How it works: Create a finite Direct-policy setup and save it as a draft. A saved preparation draft that can be reviewed before execution is considered.
Illustrative example: Draft an evaluation of support answers without sending a provider request.
Access: Builder Byok Controlled Beta. Preparation does not execute a provider or certify a Benchmark. Provider execution may require activation and your own provider API key; provider charges may apply.
Finite Direct-policy preparation and saved draft. Input: text / cross-cutting.
Plan to capture selected browser content into a dataset.
The intended browser selection-to-Dataset flow is not enabled for customers today.
What you get: No customer-ready capture result is currently offered.
Customer browser capture and reopening a dataset have not yet been qualified together.
When to use it: Consider this intended path when browser selections would be useful evaluation material.
How it works: The intended browser selection-to-Dataset flow is not enabled for customers today. No customer-ready capture result is currently offered.
Illustrative example: A team might want to collect selected example text, but the customer flow is not yet qualified.
Access: Builder Byok Controlled Beta. Customer browser capture and reopening a dataset have not yet been qualified together.
Installed browser selection-to-Dataset workflow pending. Input: text / cross-cutting.
No customer activation is offered for this path today.
Schedule a reviewed assessment of stored audio evidence.
With explicit setup, schedule evaluation of eligible stored audio evidence under the reviewed scope.
What you get: A scoped scheduled evaluation rather than general production monitoring.
Operator source, consent and scheduler controls require explicit setup; no general production monitoring claim.
When to use it: Use it when a reviewed operator-controlled audio evaluation needs repeating.
How it works: With explicit setup, schedule evaluation of eligible stored audio evidence under the reviewed scope. A scoped scheduled evaluation rather than general production monitoring.
Illustrative example: A team with the required setup schedules a review of eligible stored audio.
Access: Builder Byok Controlled Beta. Operator source, consent and scheduler controls require explicit setup; no general production monitoring claim.
Reviewed scheduled evaluation of stored audio evidence. Input: text / cross-cutting.
Evaluate submitted evidence with a bounded evaluation pack.
Submit evidence to a supported pack for a scoped evaluation.
What you get: A pack evaluation result for the submitted evidence.
Provider and agent runs remain separately activated.
When to use it: Use it when a defined pack fits the evidence you already have.
How it works: Submit evidence to a supported pack for a scoped evaluation. A pack evaluation result for the submitted evidence.
Illustrative example: Evaluate submitted examples using an eligible pack before discussing the findings.
Access: Builder Byok Controlled Beta. Provider and agent runs remain separately activated. Provider execution may require activation and your own provider API key; provider charges may apply.
Bounded submitted-evidence pack evaluation. Input: text / cross-cutting.
Connect an approved project to a scoped provider request.
Configure an exact project grant and bring your own provider API key for an eligible Gateway request.
What you get: A response to the specifically authorized request under current controls.
Provider execution remains default off and requires current controls.
When to use it: Use it when an approved application needs an eligible provider path.
How it works: Configure an exact project grant and bring your own provider API key for an eligible Gateway request. A response to the specifically authorized request under current controls.
Illustrative example: A project administrator requests activation for one provider and operation before a request is made.
Access: Builder Byok Controlled Beta. Provider execution remains default off and requires current controls. Provider execution may require activation and your own provider API key; provider charges may apply.
Exact project-scoped Gateway request after BYOK and grant. Input: text / cross-cutting.
Explore the intended path for customer image generation.
Customer image generation and its saved-result workflow are not yet enabled.
What you get: No qualified customer image result is currently offered.
Customer image generation and saved results are not yet qualified.
When to use it: Consider this future path when your evaluation needs generated images.
How it works: Customer image generation and its saved-result workflow are not yet enabled. No qualified customer image result is currently offered.
Illustrative example: A team may want to review a generated image, but this customer workflow remains unavailable.
Access: Builder Byok Controlled Beta. Customer image generation and saved results are not yet qualified. Provider execution may require activation and your own provider API key; provider charges may apply.
Customer image generation and saved result pending. Input: image / vision.
No customer activation is offered for this path today.
Use scoped native Prompt and Report APIs without provider keys.
Call the authenticated Prompt and Report APIs within an approved scope.
What you get: A native Prompt or Report result for the permitted request.
Keyless means no provider key, not anonymous access.
When to use it: Use it when an application needs an eligible native result without a provider key.
How it works: Call the authenticated Prompt and Report APIs within an approved scope. A native Prompt or Report result for the permitted request.
Illustrative example: An authorized application requests a native prompt analysis using its scoped access.
Access: Builder Byok Controlled Beta. Keyless means no provider key, not anonymous access.
Authenticated scoped native Prompt and Report APIs. Input: text / cross-cutting.
Review retained provider comparisons and availability information.
Inspect retained comparison evidence and read-only provider availability.
What you get: A view of existing provider information for a scoped comparison.
New provider discovery or execution requires separate exact authority.
When to use it: Use it when revisiting evidence already collected about providers.
How it works: Inspect retained comparison evidence and read-only provider availability. A view of existing provider information for a scoped comparison.
Illustrative example: Review a saved provider comparison before deciding whether more evidence is needed.
Access: Builder Byok Controlled Beta. New provider discovery or execution requires separate exact authority. Provider execution may require activation and your own provider API key; provider charges may apply.
Retained comparison and read-only provider availability. Input: text / cross-cutting.
Review a cited answer against its scoped retrieval evidence.
Use the bounded cited-answer retrieval path, then save and reopen the result.
What you get: A saved answer and citations tied to the tested corpus and evidence window.
Grounding applies to the tested corpus and evidence window.
When to use it: Use it when checking how a retrieval-based answer is supported.
How it works: Use the bounded cited-answer retrieval path, then save and reopen the result. A saved answer and citations tied to the tested corpus and evidence window.
Illustrative example: Review a saved answer against citations from the corpus used in its evaluation.
Access: Builder Byok Controlled Beta. Grounding applies to the tested corpus and evidence window. Provider execution may require activation and your own provider API key; provider charges may apply.
Bounded cited-answer retrieval, save and exact reopen. Input: text / vector.
Connect an eligible application using a controlled SDK package.
Obtain an administrator-delivered package and use it within current approved rights.
What you get: A scoped installed SDK integration for an eligible project.
Packages are not publicly published; administrator delivery and current rights are required.
When to use it: Use it when your approved team needs a supported integration package.
How it works: Obtain an administrator-delivered package and use it within current approved rights. A scoped installed SDK integration for an eligible project.
Illustrative example: An administrator provides a controlled package to a project team after reviewing access.
Access: Builder Byok Controlled Beta. Packages are not publicly published; administrator delivery and current rights are required.
Controlled package distribution and scoped installed SDK use. Input: text / cross-cutting.
Explore the intended path for broader multimodal execution.
Customer execution for exact provider and modality combinations is not yet qualified.
What you get: No qualified Universal customer output is currently offered.
Customer execution and saved results are not yet qualified for this path.
When to use it: Consider this intended path when planning a multimodal evaluation.
How it works: Customer execution for exact provider and modality combinations is not yet qualified. No qualified Universal customer output is currently offered.
Illustrative example: A team wants a cross-modal run, but the customer execution path is not yet available.
Access: Builder Byok Controlled Beta. Customer execution and saved results are not yet qualified for this path. Provider execution may require activation and your own provider API key; provider charges may apply.
Exact-tuple customer execution and saved result pending. Input: text / vector / image / document / audio / video / trace.
No customer activation is offered for this path today.
Understand the intended path for video evaluation.
A customer video-generation result remains unqualified and unavailable today.
What you get: No eligible customer-generated clip or video-understanding result is offered.
No eligible customer-generated clip or video-understanding result has been qualified.
When to use it: Consider this intended path when planning a future video workflow.
How it works: A customer video-generation result remains unqualified and unavailable today. No eligible customer-generated clip or video-understanding result is offered.
Illustrative example: A team may want to review a clip later, but cannot use this path for a qualified result now.
Access: Builder Byok Controlled Beta. No eligible customer-generated clip or video-understanding result has been qualified. Provider execution may require activation and your own provider API key; provider charges may apply.
Bounded video generation customer result pending. Input: video.
No customer activation is offered for this path today.
Review the scanner-first document evaluation approach.
Review a locally proven scanner-first fictional PDF OCR path for release consideration.
What you get: A review-only demonstration of the documented scanning and OCR approach.
Customer activation is withheld while the scanner and OCR release packaging and broader vision checks remain open.
When to use it: Use this review to understand the proposed document boundary before customer activation.
How it works: Review a locally proven scanner-first fictional PDF OCR path for release consideration. A review-only demonstration of the documented scanning and OCR approach.
Illustrative example: A fictional PDF is used in local review; it does not establish general customer OCR availability.
Access: Builder Byok Controlled Beta. Customer activation is withheld while the scanner and OCR release packaging and broader vision checks remain open. Provider execution may require activation and your own provider API key; provider charges may apply.
Locally proven scanner-first fictional PDF OCR for release review. Input: image / document.
No customer activation is offered for this path today.
Review submitted audio and caller-declared transcript evidence.
Provide a scanned WAV and a caller-declared transcript for the scoped evaluation.
What you get: An evaluation of the submitted audio and transcript evidence.
No independent ASR, speaker identity, voice cloning or realtime inference is claimed.
When to use it: Use it when reviewing eligible audio evidence without relying on inferred speaker identity.
How it works: Provide a scanned WAV and a caller-declared transcript for the scoped evaluation. An evaluation of the submitted audio and transcript evidence.
Illustrative example: Evaluate a provided WAV alongside its declared transcript for a scoped review.
Access: Builder Byok Controlled Beta. No independent ASR, speaker identity, voice cloning or realtime inference is claimed. Provider execution may require activation and your own provider API key; provider charges may apply.
Scanned WAV and caller-declared submitted transcript evaluation. Input: audio / realtime.
Review a declared performance band in a private pilot.
Provide declared information for a reviewed enterprise assessment with required consent and privacy controls.
What you get: A saved declared-band assessment for the pilot.
Requires enterprise provisioning, consent and privacy thresholds; not real peer-performance proof.
When to use it: Use it when a provisioned enterprise is assessing its declared performance band.
How it works: Provide declared information for a reviewed enterprise assessment with required consent and privacy controls. A saved declared-band assessment for the pilot.
Illustrative example: A participating team reviews its declared band without claiming verified peer performance.
Access: Enterprise Controlled Beta. Requires enterprise provisioning, consent and privacy thresholds; not real peer-performance proof.
Reviewed declared-band assessment and saved report pilot. Input: cross-cutting.
Plan an evaluation using an owned private endpoint.
The owned HTTPS endpoint customer path is not yet enabled.
What you get: No qualified customer endpoint run is currently offered.
No approved owned endpoint or joined customer-host connectivity exists.
When to use it: Consider this intended path when planning private inference evaluation.
How it works: The owned HTTPS endpoint customer path is not yet enabled. No qualified customer endpoint run is currently offered.
Illustrative example: A team with an owned endpoint can plan a review, but the joined customer path is not yet available.
Access: Enterprise Controlled Beta. No approved owned endpoint or joined customer-host connectivity exists. Provider execution may require activation and your own provider API key; provider charges may apply.
Owned HTTPS inference endpoint customer path pending. Input: private / customer-controlled.
No customer activation is offered for this path today.
Explore a supervised private Prompt and Report pilot.
Arrange an explicitly approved private package pilot for an eligible recipient and host.
What you get: A supervised pilot of the scoped Prompt and Report package.
Recipient, host, license custody and release recovery require explicit approval; not full Private WAO.
When to use it: Use it when an enterprise needs to evaluate this limited private deployment path.
How it works: Arrange an explicitly approved private package pilot for an eligible recipient and host. A supervised pilot of the scoped Prompt and Report package.
Illustrative example: A reviewed team assesses a private Prompt/Report package on an approved host.
Access: Enterprise Controlled Beta. Recipient, host, license custody and release recovery require explicit approval; not full Private WAO.
Supervised private Prompt/Report package pilot. Input: private / customer-controlled.
Assess a licensed private Prompt and Report engine pilot.
Request reviewed enterprise provisioning and approved distribution of the scoped engine.
What you get: A licensed Prompt/Report engine pilot for an eligible organization.
Requires enterprise provisioning and approved distribution; not public self-service.
When to use it: Use it when a team is considering the limited private engine path.
How it works: Request reviewed enterprise provisioning and approved distribution of the scoped engine. A licensed Prompt/Report engine pilot for an eligible organization.
Illustrative example: An enterprise reviews licensing and provisioning before receiving an approved pilot.
Access: Enterprise Controlled Beta. Requires enterprise provisioning and approved distribution; not public self-service.
Licensed private Prompt/Report engine pilot. Input: private / customer-controlled.
Qualification boundary
Availability varies by provider, model, operation, modality, region, account and project. Payments, checkout and unrestricted signup remain disabled. No automatic fallback, provider switching or maturity promotion is implied.