artifacts = ["/app/contract.docx"]

[task]
name = "contractbench/ip-bbb-aaa-t1-e"
description = "ContractBench: BBB Interactive, LLC v. AAA Media Group, Inc. (IP Licensing, US, single-agreement playbook), turn 1 (opening redline), attorney E"
keywords = ["legal", "contract-redlining", "docx", "tracked-changes", "llm-judge"]

[metadata]
benchmark = "ContractBench"
scenario = "ip-bbb-aaa"
agreement_type = "IP Licensing"
jurisdiction = "US"
playbook = "single-agreement"
turn = 1
side = "B"
represented_party = "BBB Interactive, LLC"
counterparty = "AAA Media Group, Inc."
author = "Reviewing Counsel (BBB Interactive, LLC)"
attorney = "E"
responds_to = ""
expert_id = 16
n_rubrics = 17
n_escalate_rubrics = 2
n_high_priority = 11
n_medium_priority = 4
n_low_priority = 2
total_weight = 43
split = "public"

[environment]
build_timeout_sec = 1200.0
cpus = 2
memory_mb = 4096
skills_dir = "/skills"

[environment.env]
ESCALATION_PREFIX = "ESCALATE:"

[agent]
timeout_sec = 3600.0

[verifier]
timeout_sec = 1200.0

[verifier.env]
JUDGE_PANEL = "anthropic/claude-opus-4-8,openai/gpt-5.5,gemini/gemini-3.5-flash"
ANTHROPIC_API_KEY = "${ANTHROPIC_API_KEY}"
OPENAI_API_KEY = "${OPENAI_API_KEY:-}"
GEMINI_API_KEY = "${GEMINI_API_KEY:-}"