@evolvingmachines/evolve 0.0.67 → 0.0.68

package/skills/evolve-evals/references/sdk-reference/methods/meta.md

@@ -48,12 +48,13 @@ TypeScript returns the typed object below. Python uses dataclasses for the docum
 
 ### CapabilityDocument: SDK field definitions
 
-`limits` reports current values for this deployment. `default_model` is a picker suggestion; managed evaluation jobs still require a model. Optional fields may be absent on older servers; Python maps missing `analyze` and `gpu_concurrency_cap` to `None`.
+`limits` reports current values for this deployment. `default_model` is a picker suggestion; managed evaluation jobs still require a model. Optional fields may be absent on older servers; Python maps missing `analyze` and `gpu_concurrency_cap` to `None`, and missing `retired_agents` to an empty list.
 
 ```ts Fields
 interface CapabilityDocument {
   schema_version: number;
   agents: AgentCapability[];
+  retired_agents?: RetiredAgent[];
   agent_registration: {
     name_pattern: string;
     max_name_length: number;
@@ -77,9 +78,14 @@ interface CapabilityDocument {
     dataset_version: StatusVocabulary;
   };
   analyze?: {
+    default_agent: string;
     default_model: string;
-    reasoning_efforts: string[];
-    models: { alias: string; model_id: string; default_reasoning_effort: string }[];
+    agents: {
+      name: string;
+      default_model: string;
+      reasoning_efforts: string[];
+      models: { alias: string; model_id: string; default_reasoning_effort: string | null }[];
+    }[];
   };
   limits: {
     job: {
@@ -139,7 +145,7 @@ The API also returns the fields below. TypeScript keeps them in the response, al
 
 ### Harnesses and models
 
-`effort_support` is `level`, `binary`, or `none`. `runnable: false` includes a reason. `supports_config` controls native settings support; `presets` lists accepted named presets. Python defaults missing `supports_config` to `False` and `presets` to an empty list. `latest_version: null` means unknown, not current.
+`agents` lists the harnesses a new job may name; a retired harness appears only in `retired_agents`, with the harness that replaces it. `effort_support` is `level`, `binary`, or `none`. `runnable: false` includes a reason. `supports_config` controls native settings support; `presets` lists accepted named presets. Python defaults missing `supports_config` to `False` and `presets` to an empty list. `latest_version: null` means unknown, not current.
 
 ```ts Fields
 interface AgentModelOption {
@@ -148,6 +154,11 @@ interface AgentModelOption {
   description: string | null;
 }
 
+interface RetiredAgent {
+  agent: string;
+  replaced_by: string;
+}
+
 interface AgentCapability {
   name: string;
   runnable: boolean;

package/skills/evolve-evals/references/sdk-reference/skills.md

@@ -22,7 +22,7 @@ import { skills, jobs } from "@evolvingmachines/evolve";
 const [skill] = await skills().upload("./my-skill");
 const job = await jobs().start({
   datasets: [{ name: "harbor-examples", version: "1.0", task_names: ["hello-world"] }],
-  agents: [{ name: "codex", model_name: "gpt-5.6-luna", skills: [skill.ref] }],
+  agents: [{ name: "codex", model_name: "gpt-6-luna", skills: [skill.ref] }],
   max_trial_spend_usd: 0.50,
   retry: { max_retries: 0 },
 });
@@ -35,7 +35,7 @@ uploaded = await skills().upload("./my-skill")
 skill = uploaded[0]
 job = await jobs().start(
     datasets=[{"name": "harbor-examples", "version": "1.0", "task_names": ["hello-world"]}],
-    agents=[{"name": "codex", "model_name": "gpt-5.6-luna", "skills": [skill.ref]}],
+    agents=[{"name": "codex", "model_name": "gpt-6-luna", "skills": [skill.ref]}],
     max_trial_spend_usd=0.50,
     retry={"max_retries": 0},
 )

package/skills/evolve-evals/references/sdk-reference/types.md

@@ -208,10 +208,11 @@ interface RetryConfig {
 }
 
 interface AnalyzeConfig {
+  agent: string;
   model_name: string;
   rubric: Rubric;
   prompt: string | null;
-  reasoning_effort: string;
+  reasoning_effort: string | null;
   sandbox_provider: EvalSandboxProvider;
   n_concurrent: number | null;
   passing: boolean;
@@ -316,8 +317,9 @@ interface CheckRow {
   name: string;
   status: CheckStatus;
   source: CheckSource;
+  agent: string;
   model_name: string;
-  reasoning_effort: string;
+  reasoning_effort: string | null;
   sandbox_provider: EvalSandboxProvider;
   org: string;
   visibility: JobVisibility;
@@ -757,6 +759,7 @@ interface TrialAnalysis {
   job_id: string;
   task_name: string;
   status: AnalysisStatus;
+  agent: string;
   model_name: string;
   reasoning_effort: string | null;
   rubric: Rubric;
@@ -788,10 +791,11 @@ interface AnalysisFailure {
 }
 
 interface AnalyzeDefaults {
+  agent: string;
   model_name: string;
   rubric: Rubric;
   prompt: string;
-  reasoning_effort: string;
+  reasoning_effort: string | null;
   sandbox_provider: EvalSandboxProvider;
 }
 ```
@@ -806,8 +810,9 @@ interface Check {
   name: string;
   status: CheckStatus;
   source: CheckSource;
+  agent: string;
   model_name: string;
-  reasoning_effort: string;
+  reasoning_effort: string | null;
   rubric: Rubric;
   prompt: string | null;
   sandbox_provider: EvalSandboxProvider;
@@ -843,10 +848,11 @@ interface TaskCheck {
 }
 
 interface CheckDefaults {
+  agent: string;
   model_name: string;
   rubric: Rubric;
   prompt: string;
-  reasoning_effort: string;
+  reasoning_effort: string | null;
   sandbox_provider: EvalSandboxProvider;
 }
 ```