feat(config)!: hard-deprecate the graders: config block — a grader is just a target

CONCEPTS.md

@@ -10,6 +10,8 @@ Shared domain vocabulary for this project — entities, named processes, and sta
 
 **Target runtime** — The placement/transport mode for invoking a target provider, such as host execution, sandbox/container placement, CLI subprocess, app-server protocol, RPC, or SDK child runner. Runtime describes how the selected agent is invoked. Advanced home/env/profile-style overlays are provider or runtime configuration details, not the authored testbed recipe. Runtime is separate from the environment that prepares files, services, and cwd.
 
+**Grader (selection)** — Not a distinct entity type. A grader is a regular target, listed under `targets` like any other and selected for a grading role via `defaults.grader`, CLI `--grader-target`, or an assertion-level target override. Authoring a separate `graders:` list is a hard error: move each entry into `targets`. This also means a grader can itself be the target of an ordinary eval (oracle/calibration checks), which a separate schema would block. Do not confuse this with **grader (method)** — an assertion `type` such as `llm-rubric`, `code-grader`, or `g-eval` that names *how* scoring works, independent of *which* target performs it; or with a rubric's `criteria`/`value`, which is the grading *prompt*. AgentV once called grader-selection "judge" (`judge_target`) before a deliberate, full rename to "grader" — do not reintroduce "judge" without a stronger reason than surface-level clarity, since peer frameworks (Margin-Lab/evals, Harbor) don't treat judge/grader as a first-class schema concept either.
+
 **Provider runtime boundary** — The process boundary between AgentV's evaluation orchestrator and the agent runtime a provider invokes. CLI-backed providers place the agent runtime outside the orchestrator; SDK providers should run through an AgentV-owned child-runner boundary when runtime teardown can threaten run artifact finalization. This boundary does not own repository/testbed setup by default.
 
 ## Evaluation Model

README.md

@@ -44,9 +44,9 @@ targets:
       api_key: "{{ env.LOCAL_OPENAI_PROXY_API_KEY }}"
       model: "{{ env.LOCAL_OPENAI_PROXY_MODEL }}"
 
-graders:
   - id: local-openai-grader
     provider: openai
+    runtime: host
     config:
       api_format: chat
       base_url: "{{ env.LOCAL_OPENAI_PROXY_BASE_URL }}"

apps/web/src/content/docs/docs/next/evaluation/running-evals.mdx

@@ -474,7 +474,9 @@ default.
 ### YAML config (`config.yaml` plus optional `config.local.yaml`)
 
 Project config and eval YAML share the same eval-definition graph for targets,
-graders, tests, defaults, and supported execution defaults. Small projects can keep that
+tests, defaults, and supported execution defaults. A grader is not a separate
+part of that graph — it is a target, listed under `targets` like any other,
+selected for a grading role via `defaults.grader`. Small projects can keep the
 graph inline:
 
 ```yaml
@@ -486,9 +488,9 @@ targets:
       command: ["codex", "app-server"]
       model: gpt-5-codex
 
-graders:
   - id: openai-grader
     provider: openai
+    runtime: host
     config:
       model: gpt-5-mini
 
@@ -511,7 +513,6 @@ directly:
 
 ```yaml
 targets: file://targets.yaml
-graders: file://graders.yaml
 tests: file://tests.yaml
 defaults: file://defaults.yaml
 ```
@@ -523,6 +524,12 @@ defaults: file://defaults.yaml
   runtime: host
   config:
     command: ["codex", "app-server"]
+
+- id: openai-grader
+  provider: openai
+  runtime: host
+  config:
+    model: gpt-5-mini
 ```
 
 ```yaml

apps/web/src/content/docs/docs/next/targets/cli-provider.mdx

@@ -21,9 +21,9 @@ targets:
     config:
       command: python agent.py --prompt {PROMPT} --out {OUTPUT_FILE}
 
-graders:
   - id: azure-grader
     provider: azure
+    runtime: host
     config:
       endpoint: "{{ env.AZURE_OPENAI_ENDPOINT }}"
       api_key: "{{ env.AZURE_OPENAI_API_KEY }}"
@@ -133,9 +133,9 @@ targets:
     config:
       command: cp fixtures/{EVAL_ID}.expected.txt {OUTPUT_FILE}
 
-graders:
   - id: azure-grader
     provider: azure
+    runtime: host
     config:
       endpoint: "{{ env.AZURE_OPENAI_ENDPOINT }}"
       api_key: "{{ env.AZURE_OPENAI_API_KEY }}"

apps/web/src/content/docs/docs/next/targets/coding-agents.mdx

@@ -15,7 +15,9 @@ settings live under `config`.
 Use `defaults.grader`, CLI `--grader` / `--grader-target`, or an
 evaluator-specific target override for LLM-based grading. Grader selection is
 separate from the coding-agent target, so target definitions do not carry a
-grader field.
+grader field. A grader is not a distinct kind of entity either — it is a
+regular target, listed under `targets` like any other, selected for a grading
+role.
 
 ```yaml
 targets:
@@ -27,9 +29,9 @@ targets:
       model: gpt-5-codex
       reasoning_effort: high
 
-graders:
   - id: openai-grader
     provider: openai
+    runtime: host
     config:
       model: gpt-5-mini
 

apps/web/src/content/docs/docs/next/targets/configuration.mdx

@@ -10,8 +10,8 @@ Targets define which agent or LLM provider to evaluate. AgentV uses one
 composable config graph across project manifests and eval files:
 
 - `.agentv/config.yaml` is the project-local discovery and composition root. It
-  can hold targets, graders, tests, defaults, execution policy, results
-  settings, and repo-local project policy.
+  can hold targets, tests, defaults, execution policy, results settings, and
+  repo-local project policy.
 - `$AGENTV_HOME/config.yaml` is the user/operator config. Use it for defaults
   that apply across projects, project registry data, default result locations,
   and provider defaults that should not be copied into each repo.
@@ -43,9 +43,9 @@ targets:
       command: ["codex", "app-server"]
       model: gpt-5-codex
 
-graders:
   - id: openai-grader
     provider: openai
+    runtime: host
     config:
       model: gpt-5-mini
 
@@ -61,6 +61,11 @@ the shorthand for the current machine, or object form when you need
 settings belong under `config`. Process-backed coding-agent providers use
 `config.command` as a non-empty argv array.
 
+A grader is not a separate kind of entity — it is a target selected for a
+grading role, either through `defaults.grader` (shown above) or an
+assertion-level target override. There is no separate `graders:` list;
+authoring one is a hard error telling you to move each entry into `targets`.
+
 ## Runtime Modes
 
 Use `runtime: host` when you want AgentV to run the target exactly as it is
@@ -128,9 +133,9 @@ targets:
       command: ["codex", "app-server"]
       model: gpt-5-codex
 
-graders:
   - id: openai-grader
     provider: openai
+    runtime: host
     config:
       model: gpt-5-mini
 
@@ -151,7 +156,6 @@ can be decomposed like this:
 
 ```yaml
 targets: file://targets.yaml
-graders: file://graders.yaml
 tests: file://tests.yaml
 defaults: file://defaults.yaml
 ```
@@ -167,6 +171,12 @@ a bare array, not an object wrapped in `targets:`:
   config:
     command: ["codex", "app-server"]
     model: gpt-5-codex
+
+- id: openai-grader
+  provider: openai
+  runtime: host
+  config:
+    model: gpt-5-mini
 ```
 
 ```yaml

apps/web/src/content/docs/docs/next/targets/custom-providers.mdx

@@ -184,9 +184,9 @@ targets:
     config:
       base_url: http://localhost:8080
 
-graders:
   - id: azure-grader
     provider: azure
+    runtime: host
     config:
       endpoint: "{{ env.AZURE_OPENAI_ENDPOINT }}"
       api_key: "{{ env.AZURE_OPENAI_API_KEY }}"

apps/web/src/content/docs/docs/next/targets/llm-providers.mdx

@@ -66,9 +66,10 @@ For smoke tests and dogfood runs against a local OpenAI-compatible proxy, keep
 the endpoint, model, and placeholder key in environment variables:
 
 ```yaml
-graders:
+targets:
   - id: local-openai-grader
     provider: openai
+    runtime: host
     config:
       api_format: chat
       base_url: "{{ env.LOCAL_OPENAI_PROXY_BASE_URL }}"

examples/features/readme-quickstart/.agentv/config.yaml

@@ -1,4 +1,3 @@
 targets: file://targets.yaml
-graders: file://graders.yaml
 tests: file://tests.yaml
 defaults: file://defaults.yaml

examples/features/readme-quickstart/.agentv/targets.yaml

@@ -3,6 +3,15 @@
   runtime: host
   config:
     api_format: chat
-    base_url: ${{ LOCAL_OPENAI_PROXY_BASE_URL }}
-    api_key: ${{ LOCAL_OPENAI_PROXY_API_KEY }}
-    model: ${{ LOCAL_OPENAI_PROXY_MODEL }}
+    base_url: "{{ env.LOCAL_OPENAI_PROXY_BASE_URL }}"
+    api_key: "{{ env.LOCAL_OPENAI_PROXY_API_KEY }}"
+    model: "{{ env.LOCAL_OPENAI_PROXY_MODEL }}"
+
+- id: local-openai-grader
+  provider: openai
+  runtime: host
+  config:
+    api_format: chat
+    base_url: "{{ env.LOCAL_OPENAI_PROXY_BASE_URL }}"
+    api_key: "{{ env.LOCAL_OPENAI_PROXY_API_KEY }}"
+    model: "{{ env.LOCAL_OPENAI_PROXY_MODEL }}"

examples/features/readme-quickstart/README.md

@@ -7,14 +7,15 @@ graph into direct field refs:
 
 ```yaml
 targets: file://targets.yaml
-graders: file://graders.yaml
 tests: file://tests.yaml
 defaults: file://defaults.yaml
 ```
 
 Each referenced file contains that field's value directly, such as a bare target
 array in `.agentv/targets.yaml` and a bare defaults object in
-`.agentv/defaults.yaml`.
+`.agentv/defaults.yaml`. A grader is not a separate kind of entity — it is a
+target listed under `targets` like any other, selected for the grading role
+via `defaults.grader`.
 
 Run it against a local OpenAI-compatible endpoint:
 

packages/core/src/evaluation/loaders/config-graph.ts

@@ -61,12 +61,6 @@ export type NormalizedTargetConfig = {
   readonly config: Record<string, unknown>;
 };
 
-export type NormalizedGraderConfig = {
-  readonly id: string;
-  readonly provider: string;
-  readonly config: Record<string, unknown>;
-};
-
 export type ConfigDefaults = {
   readonly target?: string;
   readonly grader?: string;
@@ -78,7 +72,6 @@ export type ConfigExecution = {
 
 export type ComposableConfigGraph = {
   readonly targets?: readonly NormalizedTargetConfig[];
-  readonly graders?: readonly NormalizedGraderConfig[];
   readonly tests?: readonly unknown[];
   readonly defaults?: ConfigDefaults;
   readonly execution?: ConfigExecution;
@@ -124,13 +117,15 @@ export function normalizeComposableConfigGraph(
   configPath: string,
   options: NormalizeOptions = {},
 ): ComposableConfigGraph {
+  if (rawConfig.graders !== undefined) {
+    throw new Error(
+      `Field 'graders' in ${configPath} has been removed. A grader is just a target — move each entry into 'targets' and select it via 'defaults.grader' or an assertion's target override, not a separate grader list.`,
+    );
+  }
   const graph: ComposableConfigGraph = {
     ...(rawConfig.targets !== undefined
       ? { targets: parseTargets(rawConfig.targets, `${configPath}:targets`) }
       : {}),
-    ...(rawConfig.graders !== undefined
-      ? { graders: parseGraders(rawConfig.graders, `${configPath}:graders`) }
-      : {}),
     ...(rawConfig.tests !== undefined
       ? { tests: parseArray(rawConfig.tests, `${configPath}:tests`) }
       : {}),
@@ -253,20 +248,6 @@ function parseRuntime(value: unknown, location: string): NormalizedRuntimeConfig
   throw new Error(`Invalid ${location}: use 'host' or an object with mode: host|profile|sandbox.`);
 }
 
-function parseGraders(value: unknown, location: string): readonly NormalizedGraderConfig[] {
-  return parseArray(value, location).map((entry, index) => {
-    const graderLocation = `${location}[${index}]`;
-    if (!isPlainConfigObject(entry)) {
-      throw new Error(`Invalid ${graderLocation}: grader must be an object.`);
-    }
-    const id = readRequiredString(entry.id, `${graderLocation}.id`);
-    const provider = readRequiredString(entry.provider, `${graderLocation}.provider`);
-    const config = readOptionalObject(entry.config, `${graderLocation}.config`) ?? {};
-    validateCommand(config.command, `${graderLocation}.config.command`);
-    return { id, provider, config };
-  });
-}
-
 function parseDefaults(value: unknown, location: string): ConfigDefaults {
   const defaults = readOptionalObject(value, location);
   if (!defaults) {
@@ -315,27 +296,27 @@ function parseExecution(
 }
 
 function validateDefaultSelections(graph: ComposableConfigGraph, configPath: string): void {
-  // Only validated against targets/graders defined inline in this same config
-  // document. `defaults.target`/`defaults.grader` may instead name a target
-  // defined in a separately-discovered `.agentv/targets.yaml`, which this
-  // graph has no visibility into — that case is resolved (and, on an unknown
-  // name, reported) lazily at eval-run time, the same way CLI
-  // `--grader-target` already is.
-  if (graph.defaults?.target !== undefined && graph.targets && graph.targets.length > 0) {
-    const targetIds = new Set(graph.targets.map((target) => target.id));
-    if (!targetIds.has(graph.defaults.target)) {
-      throw new Error(
-        `Invalid defaults.target in ${configPath}: '${graph.defaults.target}' does not match a configured target id.`,
-      );
-    }
+  // A grader is just a target selected for a grading role, not a separate
+  // entity — `defaults.target` and `defaults.grader` both resolve against the
+  // same `targets` pool. Only validated against targets defined inline in
+  // this same config document; either may instead name a target defined in a
+  // separately-discovered `.agentv/targets.yaml`, which this graph has no
+  // visibility into — that case is resolved (and, on an unknown name,
+  // reported) lazily at eval-run time, the same way CLI `--grader-target`
+  // already is.
+  if (!graph.targets || graph.targets.length === 0) {
+    return;
   }
-  if (graph.defaults?.grader !== undefined && graph.graders && graph.graders.length > 0) {
-    const graderIds = new Set(graph.graders.map((grader) => grader.id));
-    if (!graderIds.has(graph.defaults.grader)) {
-      throw new Error(
-        `Invalid defaults.grader in ${configPath}: '${graph.defaults.grader}' does not match a configured grader id.`,
-      );
-    }
+  const targetIds = new Set(graph.targets.map((target) => target.id));
+  if (graph.defaults?.target !== undefined && !targetIds.has(graph.defaults.target)) {
+    throw new Error(
+      `Invalid defaults.target in ${configPath}: '${graph.defaults.target}' does not match a configured target id.`,
+    );
+  }
+  if (graph.defaults?.grader !== undefined && !targetIds.has(graph.defaults.grader)) {
+    throw new Error(
+      `Invalid defaults.grader in ${configPath}: '${graph.defaults.grader}' does not match a configured target id.`,
+    );
   }
 }
 

packages/core/src/evaluation/loaders/config-loader.ts

@@ -265,7 +265,6 @@ function parseConfigObject(
       ...(refs && { refs }),
       ...(tags && { tags }),
       ...(graph.targets && { targets: graph.targets }),
-      ...(graph.graders && { graders: graph.graders }),
       ...(graph.tests && { tests: graph.tests }),
       ...(graph.defaults && { defaults: graph.defaults }),
       configDir: projectDir,

packages/core/src/evaluation/validation/eval-file.schema.ts

@@ -834,14 +834,6 @@ const ConfigTargetSchema = z
   })
   .strict();
 
-const ConfigGraderSchema = z
-  .object({
-    id: z.string().min(1),
-    provider: z.string().min(1),
-    config: JsonRecordSchema.optional(),
-  })
-  .strict();
-
 const ConfigDefaultsSchema = z
   .object({
     target: z.string().min(1).optional(),
@@ -918,7 +910,12 @@ export const EvalFileSchemaInput: z.ZodType = z.object({
     })
     .optional(),
   // Shared composable config graph fields
-  graders: z.union([z.array(ConfigGraderSchema), z.string().min(1)]).optional(),
+  graders: z
+    .never({
+      invalid_type_error:
+        "Top-level 'graders' has been removed. A grader is just a target — move each entry into 'targets' and select it via 'defaults.grader' or an assertion's target override.",
+    })
+    .optional(),
   defaults: z.union([ConfigDefaultsSchema, z.string().min(1)]).optional(),
   // Removed legacy aliases
   eval_cases: z