{"resourceId":"microsoft-foundry-copilot-harness-preview-billing-dec-1-hosted-agents-ga-20261028","versions":[{"version":"external-3551c80abad0275f21ff667a905d60c34d5e42b90f727093970df855f75523f6","resource":{"id":"microsoft-foundry-copilot-harness-preview-billing-dec-1-hosted-agents-ga-20261028","title":"Microsoft introduces the Copilot harness in Foundry Agent Service (public preview) and GA long-running resilience for hosted agents; harness billing starts Dec. 1 at per-vCPU-hour and per-GiB-hour rates (Oct 28)","organization":"Microsoft (Foundry)","sector":"Cross-SLED (developers building agents)","geography":"Global (commercial Azure)","publishedAt":"2026-10-28","publicationDate":"2026-10-28","eventDate":"2026-12-01","sourceName":"Microsoft Foundry Blog","sourceLabel":"Vendor product blog","sourceUrl":"https://devblogs.microsoft.com/foundry/microsoft-copilot-harness-in-foundry/","evidenceClass":"vendor-claim","outcomeClass":"emerging","topics":["developers-agents","infrastructure","governance-procurement"],"finding":"Microsoft's Foundry blog (Oct. 28) introduces the Microsoft Copilot harness in Foundry Agent Service in public preview: a managed execution system in which Microsoft runs the agent loop and sandbox while customers supply skills, MCP and Foundry tools (with configured authentication and approvals), files and subagents. It also announces general availability of long-running resilience for hosted agents, which restarts interrupted work with its original input while customer code handles checkpoints and duplicate-action prevention. Harness runtime is free until Dec. 1, 2026; from then sandbox compute and memory are billed at $0.0994 per vCPU-hour and $0.0118 per GiB-hour, plus model inference and tools. Customer quotes are from commercial firms, not the public sector.","sledRelevance":"Agencies and universities building agents on Azure get a managed agent runtime with a new consumption meter from Dec. 1. Government-cloud availability is not stated, so it pairs with the Azure Government feature page in this edition.","evidence":"Vendor announcement with feature description and pricing. No public-sector customers or independent performance data.","architectureImplications":"Choice between a Microsoft-managed harness and hosted agents with your own framework; long-running work needs application-level checkpointing and idempotency.","governanceImplications":"Tool approvals and policy boundaries are configurable, but agent behaviour still needs evaluation and ownership by the customer.","securityPrivacyImplications":"Sandboxed execution and MCP tool connections expand the attack surface; authentication and approval settings must be reviewed.","caveats":"Vendor claim; preview features can change and may lack compliance commitments. No statement of availability in Azure Government. Prices may vary by region and agreement.","streamIds":["microsoft"],"roles":{"sales":"This is a developer-platform announcement, so the SLED conversation is about agent ambitions and cost visibility, not a ready public-sector solution. Ask which long-running tasks an agency or university wants agents to do and whether its workloads run in commercial Azure or Azure Government. Flag that harness billing starts Dec. 1 on a per-vCPU-hour basis plus inference. Do not claim government-cloud availability or public-sector references; neither is in the source.","engineering":"Engineers should evaluate the trade-off between the managed harness and hosted agents. The harness handles the loop, context and sandbox; hosted agents keep your framework. For long-running work, the blog is explicit that your code must checkpoint progress and prevent duplicate actions on restart. Review MCP tool authentication and approval settings, confirm region and cloud availability, and estimate sandbox compute and memory costs before Dec. 1. Test in commercial Azure only if data rules allow.","delivery":"Delivery teams piloting agents should set a cost budget and monitoring before Dec. 1 billing, define evaluation criteria for agent outputs, and keep agent definitions in source control with change review. If the customer is in Azure Government, confirm availability first and plan for a gap. Assign an owner for each agent and its tools, and document approval steps for any action that changes another system."},"roleDetails":{"sales":{"customerProblem":"Public-sector developers want agents for multi-step work but need predictable cost and a supported cloud.","stakeholders":["Enterprise architect","Application development lead","CISO","Finance/FinOps"],"discoveryQuestions":["Which multi-step tasks are candidates for agents?","Is your workload in commercial Azure or Azure Government?","How will agent compute costs be budgeted after Dec. 1?","Who approves agent tool actions?"],"valueHypothesis":"A managed harness may shorten agent build time, but only where cloud availability and cost controls fit.","potentialEngagement":"Agent pilot design with cost model and cloud-availability check.","unsupportedClaims":["Availability in Azure Government","Public-sector customer results","Total cost of ownership"]},"engineering":{"fit":"Teams building agents in commercial Azure; uncertain for Azure Government.","architectureIntegration":"Declarative agents on Foundry Agent Service with MCP/Foundry tools, sandboxed execution, optional hosted agents.","prerequisites":["Foundry project and supported model","Tool authentication setup","Cost monitoring"],"constraints":["Public preview","Switching harness after creation not supported","Region/cloud availability unconfirmed"],"security":"Review MCP tool auth and approvals, sandbox isolation, and idempotency for system-changing actions.","proofOfValue":["Build one long-running agent with checkpoints and measure cost per run","Test restart without duplicate actions"]},"delivery":{"work":"Pilot agent build, cost tracking and evaluation.","dependencies":["Cloud availability","Budget approval","Evaluation dataset"],"ownership":"Application team with FinOps and security.","skillsAdoption":"Agent evaluation and skill authoring.","governanceCheckpoints":["Cost alert set before Dec. 1","Security review of tools","Evaluation before production"],"acceptanceCriteria":["Proposed: cost per task measured","Proposed: no duplicate actions in restart tests"],"risks":["Unexpected metered costs","Preview changes","Agent errors in connected systems"]}},"retrievedAt":"2026-10-31T03:08:28Z","enrichedAt":"2026-10-31T03:21:01Z","enrichmentBasis":"retrieved source","accessibilityWorkforceImplications":"Shifts developer effort from orchestration to skills and evaluation; teams need agent-evaluation skills.","procurementImplications":"A new metered line item (sandbox compute and memory) beginning Dec. 1 on top of model inference.","operatingModelImplications":"Agents defined declaratively and source-controlled alongside application code, suggesting DevOps ownership of agents.","sourceVerification":{"openedUrl":"https://devblogs.microsoft.com/foundry/microsoft-copilot-harness-in-foundry/","referenceExcerpt":"Billing for the Copilot harness in Foundry Agent Service begins December 1, 2026, with consumption-based pricing.","promptVersion":"sled-research-v4.0","model":null,"basis":"agent-reported inspection"}}}]}