Remove research post link and update code references

README.md

@@ -31,8 +31,6 @@ rubrics, and an *execution harness* for running and evaluating agents. This
 repository contains the dataset; the harness is available on
 [GitHub](https://github.com/SecondState-ai/finance-agents-benchmark).
 
-Read the research post: [Can an AI agent work through a deal data room?](https://github.com/SecondState-ai/finance-agents-benchmark/blob/9df6c2b4eabd2e866301989d03bae7be6231d589/docs/blogpost.md)
-
 ## Dataset
 
 **50 tasks · 160 documents · 231 grading criteria · One shared data room**
@@ -64,7 +62,7 @@ questions = load_dataset(
 )
 ```
 
-Follow the [walkthrough](https://github.com/SecondState-ai/finance-agents-benchmark/blob/9df6c2b4eabd2e866301989d03bae7be6231d589/docs/tutorial.md)
+Follow the [walkthrough](https://github.com/SecondState-ai/finance-agents-benchmark/blob/d464c70f54c1a1431c75687180c0e7ab8b2d976d/docs/tutorial.md)
 for installation, [API keys](.env.example), agent runs and grading. It requires
 Python 3.11+, `uv`, Docker or Podman, and model API access.
 
@@ -76,7 +74,7 @@ Python 3.11+, `uv`, Docker or Podman, and model API access.
 | [Data room](https://huggingface.co/datasets/secondstate/finance-agents-benchmark/tree/main/tasks/meridian/data-room) | Shared evidence files |
 | [Tasks](https://huggingface.co/datasets/secondstate/finance-agents-benchmark/tree/main/tasks/meridian/tasks) | Instructions and grading criteria |
 | [Release manifest](release-manifest.json) | Source revision and file hashes |
-| [Results report](https://github.com/SecondState-ai/finance-agents-benchmark/blob/9df6c2b4eabd2e866301989d03bae7be6231d589/reports/meridian/results-2026-09-27.json) | Scores, judge settings, usage and limitations |
+| [Results report](https://github.com/SecondState-ai/finance-agents-benchmark/blob/d464c70f54c1a1431c75687180c0e7ab8b2d976d/reports/meridian/results-2026-09-27.json) | Scores, judge settings, usage and limitations |
 
 Results cover four models, three trials each, on **all 50 tasks (600 answers)**.
 Task files match the evaluation; the harness system prompt has since changed.

release-manifest.json

@@ -12,8 +12,8 @@
       "sha256": "d91522377480531660dee720ae85060b297dae03dc8847c1d886691c68070e4d"
     },
     "README.md": {
-      "bytes": 3773,
-      "sha256": "328c56be395b27bf448203a83008be77841b2b9897b7886773ce017384bf36a2"
+      "bytes": 3576,
+      "sha256": "98bc238deebc068ca1edfb99f89a8ef96cf4838aa230afdc4242f9cf97bdf200"
     },
     "questions.jsonl": {
       "bytes": 24680,