Standard
  • Docs
  • Examples
  • Pricing
  • About

Get started

  • Overview
  • Conversation analytics
  • How Standard works
  • Run your first task
  • Connect
  • Join a Space

Build

  • Authoring · Sales calculation
  • Examples
  • Software release verification
  • Operation profiles
  • Integration reference

Concepts

  • Sources and customer vaults
  • Inputs and permissions
  • Accounts and billing
  • Results and lifecycle
  • Verification checks
  • Follow-on tasks
  • Cleanup

Security

  • Security model
  • Threat model and AI risk mapping
Documentation menu

Get started

  • Overview
  • Conversation analytics
  • How Standard works
  • Run your first task
  • Connect
  • Join a Space

Build

  • Authoring · Sales calculation
  • Examples
  • Software release verification
  • Operation profiles
  • Integration reference

Concepts

  • Sources and customer vaults
  • Inputs and permissions
  • Accounts and billing
  • Results and lifecycle
  • Verification checks
  • Follow-on tasks
  • Cleanup

Security

  • Security model
  • Threat model and AI risk mapping
Standard

Managed confidential execution for approved functions.

© 2026 Standard Runtime, Inc. All rights reserved.

  • Security
  • Docs
  • Examples
  • Pricing
  • FAQ
  • About
  • Contact
  • Terms
  • Privacy
  • Cookies
  • Docs
  • Examples
  • Pricing
  • About
  • Get started

    • Overview
    • Conversation analytics
    • How Standard works
    • Run your first task
    • Connect
    • Join a Space

    Build

    • Authoring · Sales calculation
    • Examples
    • Software release verification
    • Operation profiles
    • Integration reference

    Concepts

    • Sources and customer vaults
    • Inputs and permissions
    • Accounts and billing
    • Results and lifecycle
    • Verification checks
    • Follow-on tasks
    • Cleanup

    Security

    • Security model
    • Threat model and AI risk mapping

    Get started

    • Overview
    • Conversation analytics
    • How Standard works
    • Run your first task
    • Connect
    • Join a Space

    Build

    • Authoring · Sales calculation
    • Examples
    • Software release verification
    • Operation profiles
    • Integration reference

    Concepts

    • Sources and customer vaults
    • Inputs and permissions
    • Accounts and billing
    • Results and lifecycle
    • Verification checks
    • Follow-on tasks
    • Cleanup

    Security

    • Security model
    • Threat model and AI risk mapping

    Conversation analytics

    The conversation-categories example: protected input, duplicate-ID validation, permitted counts, recipient verification and publication policy.

    Example · Conversation analytics

    An Analytics agent requests category counts from an approved batch of 1,000 private conversations. The approved function, conversation-categories version 1, processes batch-042 inside Standard’s protected runtime. The Analytics integration verifies the permitted result before making the counts available to the agent. The reporting workflow applies its own publication policy.

    These figures are examples, not a retained execution. The classifier and any model it uses must execute inside the protected runtime. An external model API would be a separate recipient outside that boundary.

    Function and source

    Download the JavaScript reference function. It contains the classifier and duplicate-ID check used by the example tests. It is executable reference logic, not a deployed or Wasm-packaged workload. Adapt and build it for the supported Wasm interface, test the resulting artifact, and approve its exact revision and output contract before protected execution.

    The reference uses simple keyword rules with this priority: Coding, Writing, Research, then Other. Coding matches code, coding, debug or programming; Writing matches write, writing, edit or draft; Research matches research, study, analyze or sources. Matching ignores case and uses word boundaries. These rules explain the reference, not classification accuracy. Your team defines suitable categories, logic and any compatible model.

    The function has no remote model calls or other I/O. The protected Wasm function receives declared inputs and outputs under enforced execution limits; it has no general shell, filesystem, subprocess or network interface.

    Protected input

    The owner approves the connected source, selected resource, version and required data bindings. Confidential ingress retrieves the authorized batch and applies the required digest and size checks before passing it to the classifier. The Analytics agent passes the approved source reference; raw conversations and source credentials are not supplied to its environment.

    The following two synthetic records show the input format only. They are not the 1,000-conversation batch behind the example counts.

    json
    {
      "conversations": [
        { "id": "conversation-001", "text": "Help debug this code." },
        { "id": "conversation-002", "text": "Help draft a letter." }
      ]
    }

    The reference accepts at most 1,000 records. IDs must be nonempty strings of at most 128 characters, and text must be a string of at most 8,192 characters. IDs must be unique by exact string equality. These are example workload checks; runtime byte and resource limits are configured separately.

    Validate before classifying

    The function validates the complete batch before counting any category. A repeated conversation ID produces this permitted failure report, without partial counts or conversation contents:

    PreviousOverviewNextHow Standard works

    On this page

    • Function and source
    • Protected input
    • Validate before classifying
    • Permitted result
    • Execution setup
    • Request, delivery and verification
    • Verification and publication
    • Related examples and boundaries
    json
    {
      "outcome": "input_check_failed",
      "reason": "duplicate_conversation_ids"
    }

    The duplicate-ID check belongs to this function. Standard does not automatically detect duplicate conversation IDs for every workload. The approved output contract must permit both aggregate-count results and redacted failure reports.

    Permitted result

    For the example batch, the classifier produces:

    json
    {
      "outcome": "completed",
      "counts": {
        "coding": 420,
        "writing": 310,
        "research": 190,
        "other": 80
      },
      "total": 1000
    }

    The result contains no conversation text or source credentials. Counts can still reveal information, especially across repeated requests. Set the output policy and permitted queries accordingly; aggregation alone is not a complete privacy guarantee.

    Execution setup

    This example explains a configured operation; it is not a ready-to-deploy project or a claim of a completed customer transaction. Follow Run your first task and the assisted setup checklist. Hosted continuation currently refuses hosted_continuation_unavailable after admission approval when Control has reported no progress. Confirm the supported execution route before requesting the example.

    Request, delivery and verification

    1. The Analytics agent requests conversation-categories for the approved source reference batch-042 using its configured integration.
    2. The input owner approves this operation’s function, batch and permitted-result release to the Analytics integration. Connected-source work requires its per-operation authority.
    3. Standard retrieves and validates the batch, runs the approved classifier in the confidential runtime, and releases the permitted result with evidence.
    4. The hosted response reports operation and delivery metadata. It is not verified plaintext counts.
    5. The Analytics integration retrieves the result and verifies the expected function, input, authority, runtime, recipient and result against trusted references.
    6. Only an accepted result becomes available to the Analytics agent. The reporting workflow separately decides whether publication is permitted.

    This is an operation sequence, not CLI commands or SDK signatures. Use the tools and schemas your authenticated account exposes. See Integration reference.

    Verification and publication

    StateEvidenceFunction outcomeReporting policy
    Counts readyAccepted against the receiving integration’s referencesClassification completed; category counts availablePublication permitted
    Input check failedAccepted, including the failure reportDuplicate conversation IDs; no category countsReport update held
    Publication heldAccepted against the receiving integration’s referencesClassification completed; category counts availablePublication approval required
    Invalid evidenceRejected because the received result or another bound field does not matchNot accepted from this recordReport update blocked

    Verify the evidence. Evaluate the result. Enforce the policy for the next action. Standard Proof does not grant publication permission or certify classification accuracy or the truth of the conversations.

    Related examples and boundaries

    • Try the homepage comparison
    • Threat model and AI risk mapping
    • Verification checks
    • Software release verification