docs: changelog + M27 scoring & replay updates

billing/plans.mdx

@@ -21,7 +21,7 @@ Get started with no commitment.
 Credit is deducted only for **completed runs**. Failed runs are free -- you are never charged for work that did not produce a result.
 
 <Note>
-  Coming soon: runs scoring below Bronze (score &lt; 40) will be automatically marked non-billable. You will never pay for low-quality output.
+  Coming soon: runs scoring below Bronze (score &lt; 40) will be automatically marked non-billable. You will never pay for low-quality output. Follow the [changelog](/changelog) for updates.
 </Note>
 
 ## Scale (Pay as you go)

changelog.mdx

@@ -0,0 +1,44 @@
+---
+title: "Changelog"
+description: "Stay up to date with UseZombie product updates, new features, and improvements."
+---
+
+<Tip>
+  UseZombie is in **Early Access Preview**. Features below are live in the current release. APIs and agent behavior may evolve before GA.
+</Tip>
+
+<Update label="April 5, 2026" tags={["New releases", "Improvements"]}>
+  ## Resource efficiency scoring
+
+  Agent runs are now scored on actual memory and CPU usage. Agents that stay within their resource limits score higher. Agents that max out memory or hit CPU caps score lower — giving you visibility into wasteful runs.
+
+  ## Score formula v2
+
+  The scoring formula has been updated to v2. The resource axis now uses real data instead of a fixed value. All other axes are unchanged and existing scores are preserved.
+</Update>
+
+<Update label="March 30, 2026" tags={["New releases"]}>
+  ## Live run streaming (API)
+
+  The SSE stream endpoint is live: `GET /v1/runs/{id}:stream` emits gate results in real time as the agent works. CLI support (`--watch`) is coming in a future release.
+
+  ## Run replay (API)
+
+  Replay any finished run step by step via the API: `GET /v1/runs/{id}:replay` returns a structured gate narrative with exit codes, stdout/stderr, and wall time. CLI support (`zombiectl runs replay`) is coming in a future release.
+</Update>
+
+<Update label="March 28, 2026" tags={["New releases"]}>
+  ## Per-run cost control
+
+  Set token budgets, wall-time limits, and repair loop caps on each run. Runs that exceed limits are cancelled automatically.
+</Update>
+
+<Update label="March 25, 2026" tags={["New releases"]}>
+  ## OpenAPI spec
+
+  A complete OpenAPI 3.1 specification covering all 43 API endpoints is now published.
+
+  ## `@usezombie/zombiectl` on npm
+
+  The CLI is now available as a scoped npm package.
+</Update>

docs.json

@@ -62,6 +62,12 @@
               "cli/flags",
               "cli/configuration"
             ]
+          },
+          {
+            "group": "Updates",
+            "pages": [
+              "changelog"
+            ]
           }
         ]
       },
@@ -202,6 +208,10 @@
   },
   "navbar": {
     "links": [
+      {
+        "label": "Changelog",
+        "href": "/changelog"
+      },
       {
         "label": "Talk to us",
         "href": "https://usezombie.com/contact/sales"

runs/scorecard.mdx

@@ -14,7 +14,7 @@ The score is a weighted sum of four axes:
 | Completion | 40% | Did the agent finish the spec? |
 | Error rate | 30% | Ratio of passing stages to total stages |
 | Latency | 20% | Wall time compared to workspace rolling p50 baseline |
-| Resource efficiency | 10% | Compute and memory usage (stubbed until Firecracker) |
+| Resource efficiency | 10% | Memory and CPU usage relative to sandbox limits |
 
 ### Completion (40%)
 
@@ -36,8 +36,15 @@ The baseline is computed from the last 20 completed runs in the same workspace.
 
 ### Resource efficiency (10%)
 
+Measures how much memory and CPU the agent consumed during execution. An agent that does its job without hogging resources scores higher.
+
+- **Memory (70% of this axis)** — how much of the available memory the agent used. Using very little scores close to 100. Maxing out the limit scores 0.
+- **CPU (30% of this axis)** — how often the agent was slowed down by CPU limits. Running freely scores 100. Being constantly throttled scores 0.
+
+This gives you early visibility into resource-heavy agents before they become a cost problem.
+
 <Info>
-  Resource efficiency scoring is stubbed at a flat 50/100 until Firecracker-based sandboxing is available. Once sandboxing ships, this axis will measure actual CPU and memory consumption.
+  When resource data is not available (for example, during local development), this axis defaults to 50.
 </Info>
 
 ## Tier labels
@@ -78,13 +85,14 @@ curl https://api.usezombie.com/v1/runs/run_01JQ7K.../score \
 ```json
 {
   "run_id": "run_01JQ7K...",
-  "score": 87,
+  "score": 89,
   "tier": "Gold",
+  "formula_version": "2",
   "axes": {
     "completion": 100,
     "error_rate": 85,
     "latency": 72,
-    "resource_efficiency": 50
+    "resource_efficiency": 93
   }
 }
 ```

runs/troubleshooting.mdx

@@ -78,10 +78,13 @@ When a run fails or behaves unexpectedly, start here. The most common issues and
   </Accordion>
 </AccordionGroup>
 
-## Run replay (future)
+## Run replay
 
-<Info>
-  Run replay is planned but not yet available.
-</Info>
+<Note>`zombiectl runs replay` is planned for a future release. Use the API endpoint below in the meantime.</Note>
 
-When shipped, `zombiectl runs replay <id>` will produce a structured narrative of everything that happened during a run — one section per gate, one entry per loop iteration, with the full captured output at each step.
+Replay the complete event history of a finished run via the API:
+
+```bash
+curl https://api.usezombie.com/v1/runs/run_01JQ7K...:replay \
+  -H "Authorization: Bearer $ZOMBIE_TOKEN"
+```