chore(release): agent-eval 0.100.3

CHANGELOG.md

@@ -4,10 +4,11 @@ All notable changes to `@tangle-network/agent-eval` and its sibling `agent-eval-
 
 ---
 
-## [Unreleased] — eval fixture UX + manifest-correct campaign resume
+## [0.100.3] — 2026-07-01 — product benchmark contract + eval fixture UX
 
 ### Added
 
+- Published the `@tangle-network/agent-eval/product-benchmark` subpath so product agents can share one strict product-benchmark manifest, record, artifact, and integrity validator instead of copying Agent Lab or product-local schema code.
 - **Vercel-style eval fixture loading in `/campaign`.** `discoverEvalFixtures`, `loadEvalFixture`, `loadEvalFixtureScenarios`, and `planEvalFixtureRun` let agents use the simple `evals/<name>/PROMPT.md + EVAL.ts + package.json` shape while still executing through the existing `runCampaign` primitive.
 - **Dry-run planning for campaigns.** `planCampaignRun` reports `totalCells`, `cellsCached`, `cellsToRun`, per-cell cache paths, and miss reasons before any agent work starts. This is the cheap proof before spending tokens.
 - **`dispatchRef` on `runCampaign`.** Callers can include the model/tool/prompt/runtime identity in the manifest when the same dispatch function name can run different behavior.

clients/python/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "hatchling.build"
 
 [project]
 name = "agent-eval-rpc"
-version = "0.100.2"
+version = "0.100.3"
 description = "Python RPC client for @tangle-network/agent-eval — judge content against rubrics over HTTP or stdio RPC. Eval logic runs in the Node runtime; this package is a thin wire client."
 readme = "README.md"
 requires-python = ">=3.10"

clients/python/src/agent_eval_rpc/__init__.py

@@ -58,7 +58,7 @@
 try:
     __version__ = version("agent-eval-rpc")
 except PackageNotFoundError:
-    __version__ = "0.100.2"
+    __version__ = "0.100.3"
 
 __all__ = [
     "Client",

package.json

@@ -1,6 +1,6 @@
 {
   "name": "@tangle-network/agent-eval",
-  "version": "0.100.2",
+  "version": "0.100.3",
   "description": "Evaluate and improve AI agents from runs, traces, judges, and feedback. Compare candidates, cluster failures, measure lift, and gate releases.",
   "homepage": "https://github.com/tangle-network/agent-eval#readme",
   "repository": {