docs: add Platform Credits page and three-bucket model

src/content/docs/support-and-community/plans-and-billing/credits.mdx

@@ -12,6 +12,14 @@ Any interaction with Warp's Agent consumes credits. Credits are primarily based
 
 Credits also include a small hosting fee, charged only when running agents in the cloud, hosted on Warp's infrastructure. For details on cloud agent credits, see [Cloud Agent Credits](/support-and-community/plans-and-billing/credits/#cloud-agent-credits).
 
+Warp meters credits across three types of infrastructure — inference, compute, and platform — each with its own bucket:
+
+* **AI credits** - Charged for inference infrastructure Warp pays for. Consumed by Agent conversations and other AI features that route through Warp-managed model providers.
+* **Cloud Agent Credits** - Charged for compute infrastructure when cloud agents run on Warp-hosted compute. See [Cloud Agent Credits](/support-and-community/plans-and-billing/credits/#cloud-agent-credits).
+* **Platform Credits** - Charged for Warp's platform infrastructure (run lifecycle, integrations, dashboard, APIs, and observability). Applies to every cloud agent run, plus local agent runs on Business and Enterprise plans that use customer-supplied inference (BYOK or BYOLLM via Amazon Bedrock or Google Vertex). See [Platform Credits](/support-and-community/plans-and-billing/platform-credits/).
+
+All three buckets draw from the same plan credit pool and [Add-on Credits](/support-and-community/plans-and-billing/add-on-credits/), so you can manage them as a single balance in **Settings** > **Billing and usage**.
+
 Each interaction consumes **at least one credit**, though more complex interactions may use **multiple credits**. Because of factors such as codebase size, model choice, number of tool calls, and the nature of LLMs, credit usage is **non-deterministic** — two similar prompts can still use a different number of credits.
 
 :::note
@@ -130,3 +138,25 @@ The following scenarios do **not** use Cloud Agent Credits:
 
 * **Local agent runs** — Using `oz agent run` on your local machine
 * **Self-hosted compute** — Using `oz agent run` on GitHub Actions, CI/CD pipelines, or other self-hosted infrastructure
+
+### Platform Credits
+
+Platform Credits charge for Warp's platform infrastructure — run lifecycle, scheduling, integrations, dashboard, APIs, and observability — for every cloud agent run, plus local agent runs on Business and Enterprise plans that use customer-supplied inference.
+
+#### Eligible for Platform Credits
+
+The following scenarios use Platform Credits:
+
+* **Cloud agents on any plan** — Every cloud agent run, regardless of harness or inference source
+* **Local agents on Business or Enterprise with customer-supplied inference** — Local agent runs on a Business or Enterprise plan that use [Bring Your Own API Key (BYOK)](/support-and-community/plans-and-billing/bring-your-own-api-key/) or BYOLLM via Amazon Bedrock or Google Vertex
+
+#### Not eligible for Platform Credits
+
+The following scenarios do **not** use Platform Credits:
+
+* **Local agents on Free, Build, or Max plans** — Regardless of inference source
+* **Local agents on Business or Enterprise using Warp-managed inference** — AI credits already cover the cost
+* **Regular terminal usage** — Shell commands and non-AI features
+* **Third-party harnesses run directly** — Running `claude`, `codex`, or another CLI outside of Oz
+
+For a full breakdown of how Platform Credits work, see [Platform Credits](/support-and-community/plans-and-billing/platform-credits/).

src/content/docs/support-and-community/plans-and-billing/plans-pricing-refunds.mdx

@@ -20,6 +20,7 @@ Visit [warp.dev/pricing](https://warp.dev/pricing) to see the latest plans and w
 
 * [Credits](/support-and-community/plans-and-billing/credits/) — learn how credits are used and calculated across AI features.
 * [Add-on Credits](/support-and-community/plans-and-billing/add-on-credits/) — purchase additional credits or enable automatic reloads at discounted rates.
+* [Platform Credits](/support-and-community/plans-and-billing/platform-credits/) — the third credit bucket, which covers Warp's platform infrastructure for every cloud agent run and for local Business and Enterprise runs that use customer-supplied inference.
 * [Bring Your Own API Key](/support-and-community/plans-and-billing/bring-your-own-api-key/) — connect your own model provider API keys for custom usage and billing.
 * [Overages (Legacy)](/support-and-community/plans-and-billing/overages-legacy/) — information for users on legacy plans with overages enabled.
 * [Pricing FAQs](/support-and-community/plans-and-billing/pricing-faqs/) — answers to common questions about plans, billing, and usage. Don’t have Warp yet? [Download Warp](https://warp.dev/download) and get started for free today.

src/content/docs/support-and-community/plans-and-billing/platform-credits.mdx

@@ -0,0 +1,60 @@
+---
+title: Platform Credits
+description: >-
+  Platform Credits charge for Warp's platform infrastructure on every cloud
+  agent run, plus local agent runs on Business and Enterprise plans that use
+  customer-supplied inference. Learn when they apply and where they appear.
+---
+
+Warp meters credits across three types of infrastructure: **inference** (the model call), **compute** (the sandbox an agent runs in), and **platform** (everything that runs around the agent — run lifecycle, integrations, dashboard, APIs, and observability). Each type maps to one credit bucket: [AI credits](/support-and-community/plans-and-billing/credits/), [Cloud Agent Credits](/support-and-community/plans-and-billing/credits/#cloud-agent-credits), and Platform Credits.
+
+Platform Credits charge for Warp's platform infrastructure on every cloud agent run, plus local agent runs on Business and Enterprise plans that use customer-supplied inference. Even when you bring your own model provider, connect your own LLM API keys, or run cloud agents on your own compute, Warp's platform still coordinates, observes, and integrates your agents — and Platform Credits charge for that.
+
+Customer-supplied inference covers [Bring Your Own API Key (BYOK)](/support-and-community/plans-and-billing/bring-your-own-api-key/) and BYOLLM via Amazon Bedrock or Google Vertex. Platform Credits don't apply to local runs on Free, Build, or Max plans, or to local Business and Enterprise runs where Warp pays for inference through AI credits.
+
+## The three credit buckets
+
+Each credit bucket charges for a different layer of the infrastructure Warp provides:
+
+* **AI credits** - Charged for inference infrastructure Warp pays for. Consumed by Agent conversations, [Generate](/agent-platform/local-agents/generate/), [AI Autofill](/knowledge-and-collaboration/warp-drive/workflows/#ai-autofill), and other AI features that route through Warp-managed model providers. See [credits](/support-and-community/plans-and-billing/credits/) for how AI credits are calculated.
+* **Cloud Agent Credits** - Charged for compute infrastructure when cloud agents run on Warp-hosted compute. Consumed by Slack and Linear integrations, `oz agent run-cloud`, the Oz API, and Cloud Mode in the Warp app. See [Cloud Agent Credits](/support-and-community/plans-and-billing/credits/#cloud-agent-credits).
+* **Platform Credits** - Charged for Warp's platform infrastructure (run lifecycle, integrations, dashboard, APIs, and observability). Applies to every cloud agent run, plus local agent runs on Business and Enterprise plans that use customer-supplied inference.
+
+The three buckets are independent and a single run can consume from more than one. A Warp-managed cloud agent run, for example, consumes AI credits for the model call, Cloud Agent Credits for the hosted compute, and Platform Credits for the platform infrastructure that runs the agent.
+
+## How Platform Credits are measured
+
+Platform Credits accrue while an Oz agent is actively working on a task — from when the task starts running to when it finishes, fails, or is cancelled. Idle time, time spent waiting on user input, and time before the task starts don't accrue Platform Credits.
+
+See [warp.dev/pricing](https://warp.dev/pricing) for current rates.
+
+## When Platform Credits apply
+
+Whether Platform Credits are charged depends on where the agent runs and who's paying for inference.
+
+### Charged
+
+* **Cloud agents on any plan** - Every cloud agent run charges Platform Credits, regardless of harness (Warp Agent, Claude Code, Codex, Gemini), inference source (Warp-managed, BYOK, or BYOLLM), or compute source (Warp-hosted or self-hosted workers). Warp's platform infrastructure is always running the agent.
+* **Local agents on Business or Enterprise with customer-supplied inference** - Local agent runs on a Business or Enterprise plan that use [BYOK](/support-and-community/plans-and-billing/bring-your-own-api-key/) or BYOLLM via Amazon Bedrock or Google Vertex. Warp isn't paying for the model call, but Warp's platform infrastructure is still running the local agent.
+
+### Not charged
+
+* **Local agents on Free, Build, or Max plans** - No Platform Credits charge, regardless of whether you use Warp-managed inference or BYOK.
+* **Local agents on Business or Enterprise using Warp-managed inference** - Warp is already paying for the model call through AI credits, so no separate Platform Credits charge applies.
+* **Regular terminal usage** - Running shell commands and using non-AI Warp features never consumes Platform Credits.
+* **Third-party harnesses run directly** - Running `claude`, `codex`, or another harness as a standalone CLI without going through Oz. Oz isn't in the path, so there's nothing to charge.
+
+## Where Platform Credits appear
+
+Platform Credits roll into the same surfaces as your other credit usage.
+
+* **Per-user credit totals** - Your total credit usage in **Settings** > **Billing and usage** in the Warp app includes Platform Credits alongside AI and Cloud Agent Credits.
+* **Admin usage breakdown** - For team admins, the Admin Panel billing view shows a per-bucket breakdown for each member and for the team as a whole. Platform Credits appear as their own segment in the stacked usage bar, distinct from the AI and Compute segments.
+* **Add-on Credits and limits** - Platform Credits draw from the same plan credits and [Add-on Credits](/support-and-community/plans-and-billing/add-on-credits/) pools as your other usage. Once your monthly plan credits are exhausted, Platform Credits continue to consume from Add-on Credits if you have them.
+
+## Related pages
+
+* [Credits](/support-and-community/plans-and-billing/credits/) - How AI credits are measured and calculated.
+* [Add-on Credits](/support-and-community/plans-and-billing/add-on-credits/) - Purchase additional credits or enable auto reload.
+* [Bring Your Own API Key](/support-and-community/plans-and-billing/bring-your-own-api-key/) - Connect your own model provider keys.
+* [Plans, pricing, and refunds](/support-and-community/plans-and-billing/plans-pricing-refunds/) - Compare plans and refund policies.