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

    Follow-on tasks

    Use one task's verified result as the input to another, with evidence linking the two.

    PreviousVerification checksNextCleanup

    On this page

    • Configure the receiving step
    • Example: verified counts to a report
    • Preserve the evidence boundary

    A dependent step must establish that the result it uses came from the required operation. Configure its trusted integration to verify Standard Proof against accepted references, evaluate the result, and enforce authority for the next action before proceeding. These requirements apply when the same agent continues its work and when one agent hands work to another.

    Configure the receiving step

    1. Identify the result and the evidence the next operation requires.
    2. Verify that evidence against accepted references and policy.
    3. Apply the receiving step's acceptance rules to the result.
    4. Confirm authority to use that result and request the next operation.

    Receiving a verified result does not grant access to the original inputs. A follow-on operation needs authority for any additional resources or disclosure.

    Example: verified counts to a report

    The Analytics integration verifies the counts produced by conversation-categories for batch-042. The reporting workflow then checks its publication policy before updating the report. The analytics agent receives the permitted counts, not the protected conversations.

    Invalid evidence blocks the handoff. A verified duplicate-ID failure report contains no counts, so the report update is held. Valid counts can also remain held pending publication approval. A changed batch needs its own applicable operation and authority. See Conversation analytics.

    The separate software release verification example applies the same sequence to a candidate artifact and deployment policy.

    Preserve the evidence boundary

    Keep the upstream result and evidence references when the supported integration provides them. Claim a linked execution history only for relationships actually recorded and checked. A chain of accepted records does not establish the truth of unverified inputs or the correctness of every business decision.

    If an agent summarizes, changes, or combines a verified result, the original proof still covers only the original result. Newly claimed execution or transformation needs its own evidence.

    Each agent may use a different workflow or belong to a different owner. The verification and authority requirements apply at each handoff and at each step one agent takes on its own earlier result.