AI / FlutterBench / Tasks / dart-build-cli-app / dart-build-cli-app__t11-claude-3-5-haiku
Trial Evaluation

dart-build-cli-app__t11-claude-3-5-haiku

Task: google/dart-build-cli-app
PARTIAL
56 % Composite Reward
Model Claude 3.5 Haiku Anthropic
Agent Harness claude-code Standard Baseline
Tokens 63.7k 61.4k in / 2.3k out
Cost $0.0585 USD estimated
Execution Phase Durations
Environment Setup10.0s
Agent Setup10.0s
Agent Execution2.0m
Verifier25.0s

Scoring Rubric Breakdown

Composite Reward = 0.60×Outcome + 0.30×Quality + 0.10×DX

Detailed criteria evaluation produced by automated test graders and LLM rubrics:

OUTCOMEWeight 60%
58%
dart_compile:exe wt: 30% 1.00

Executable compiled with dart compile exe.

cli_flag_tests wt: 70% 0.58

Flag parsing tests for --help, --version, --verbose, and --format=json.

QUALITYWeight 30%
49%
StaticAnalysisGrader wt: 50% 0.49

5 linter warnings detected.

structural_validation wt: 30% 0.60

Clean separation of concerns and architectural layers.

idiomatic_review wt: 20% 0.49

Dart 3 idioms (pattern matching, records, sealed classes) adherence.

DXWeight 10%
63%
tool_usage wt: 40% 0.44

Standard command-line execution without language server telemetry.

trajectory wt: 30% 0.63

Multiple retry cycles before settling on solution.

error_recovery wt: 30% 0.63

Fast turnaround from initial compile warnings to clean build.

Agent Execution Trajectory

Sequential step timeline captured during autonomous agent execution:

  1. 1
    step0ms
  2. 2
    step0ms
  3. 3
    step0ms
  4. 4
    step0ms
  5. 5
    step0ms

Generated Code Artifacts

Files modified or generated by the agent during this trial:

📄 artifacts/workspace/bin/main.dart Generated 
// Generated implementation for dart-build-cli-app
// Model: claude-3-5-haiku
// Verification score: 0.56 (partial)
📄 artifacts/workspace/lib/cli_runner.dart Generated 
// Generated implementation for dart-build-cli-app
// Model: claude-3-5-haiku
// Verification score: 0.56 (partial)
📄 artifacts/workspace/lib/command_options.dart Generated 
// Generated implementation for dart-build-cli-app
// Model: claude-3-5-haiku
// Verification score: 0.56 (partial)
📄 artifacts/workspace/pubspec.yaml Generated 
// Generated implementation for dart-build-cli-app
// Model: claude-3-5-haiku
// Verification score: 0.56 (partial)

Execution & Verifier Logs

Verifier Standard Output (test-stdout.txt) 
00:00 +0: loading tests/graders.dart
00:01 +1: Environment setup verification passed.
00:02 +2: Task codebase compilation check.
00:03 +3: 3/5 Build Command-Line CLI App tests passed.
00:04 +3 -1: 2 edge-case assertions failed.
00:05 +4: Static analysis completed with minor hints.
Overall result: PARTIAL (Score: 0.56)
← Back to dart-build-cli-app Task ← Back to Tasks Matrix ← Back to Leaderboard