diff --git a/README.md b/README.md
index 8567f45..1fb2841 100644
--- a/README.md
+++ b/README.md
@@ -167,6 +167,7 @@ Most "awesome" lists are link dumps. This one is **annotated and verified**: eve
 - **[Harbor](https://github.com/harbor-framework/harbor)** — harbor-framework (Laude Institute / Stanford) — <https://github.com/harbor-framework/harbor> — 🆕 framework for running agent evals + creating/using RL environments; powers Terminal-Bench 2.0. ~3.7k★. ⚠️ name overloaded (cf. `av/harbor` local-LLM toolkit).
 - **[Caliper](https://github.com/edonadei/caliper)** — Emrick Donadei — <https://github.com/edonadei/caliper> · <https://pypi.org/project/caliper-eval/> — 🆕 **pass@k reliability harness for agent skills**: runs a skill k times against Claude Code / Codex / pi, grading each attempt with an LLM-autorater (`expect:`) and/or deterministic Python (`assert:`), then a `--baseline` re-run without the skill reports the delta to prove the skill beats the base agent. Git-diffable `.eval.yaml` specs; isolated per-attempt sandboxes.
 - **[Coder Eval](https://github.com/UiPath/coder_eval)** — UiPath — <https://github.com/UiPath/coder_eval> · <https://coder-eval.com/docs> — 🆕 treats **skill activation as a measurable variable**: `skill_triggered` over a labelled dataset yields suite-level precision/recall/F1 gates (cf. §3). Harness-level, not leaderboard-level. ~0.1k★.
+- **[AgentRunProof](https://github.com/FU-max-boop/agentrunproof)** — Fu Xiaonan — <https://github.com/FU-max-boop/agentrunproof> · <https://github.com/FU-max-boop/agentrunproof/blob/main/docs/case-study-runstate.md> · *tool/repo* — 🆕 Provider-free OpenAI Agents runtime-regression harness: deterministic scripts drive the real `Runner`, compare `run` / `run_streamed` and `RunState` resume invariants, and emit content-addressed records; the case study follows the public evidence chain through merged upstream fixes [#4413](https://github.com/openai/openai-agents-python/pull/4413) and [#4414](https://github.com/openai/openai-agents-python/pull/4414). ⚠️ Tests SDK runtime correctness, not model quality; independent project, not OpenAI adoption or endorsement.
 
 ### 5b · TypeScript/JS-native eval runners
 - **[evalite](https://github.com/mattpocock/evalite)** — Matt Pocock — <https://github.com/mattpocock/evalite> — 🆕 local-first eval runner on Vitest; `.eval.ts` files, web UI, cost-aware.