Company-neutral workflow kit for creating, reviewing, calibrating, and packaging Harbor / Terminal-Bench tasks with LLM agents.
-
Updated
May 18, 2026 - Python
#
benchmark-tasks
Here are 7 public repositories matching this topic...
Rebuild a Rust crate at a historical commit in a digest-pinned Docker image and verify a fail-to-pass flip
-
Updated
Sep 30, 2026 - Python
An offline benchmark task using Terraform, Python, and Graphviz to reconstruct corrupted machine learning model lineage metadata, query Hugging Face dataset split counts, and render color-coded lineage graphs.
python docker graphviz terraform pytest data-recovery huggingface-datasets lineage-tracking benchmark-tasks
-
Updated
Jul 20, 2026 - Python
Mine and rank candidate fail-to-pass tasks from real repository history: file classifier, transparent scorer, patch-fingerprint dedupe ledger, JSONL export.
-
Updated
Sep 30, 2026 - Python
An AI coding benchmark task evaluating automated rebase conflict resolution, MLflow tracking security, secret redaction, and Hugging Face model evaluation safety.
python3 git-rebase cybersecurity security-hardening mlops mlflow huggingface agent-evaluation agent-eval benchmark-tasks
-
Updated
Jul 25, 2026 - Python
Review gates for pull requests that add or change benchmark tasks for AI coding agents: Python CLI and composite GitHub Action, with Docker runner, determinism reruns and a content-hash cache
-
Updated
Sep 30, 2026 - Python
An automated media verification & reconciliation engine resolving conflicting SQLite attestations, Ed25519 signed release manifests, and lost Git verifier policies via Git reflog recovery.
python cryptography reconciliation sqlite pytest ed25519 attestation audit-trail digital-watermark git-reflog git-recovery benchmark-tasks
-
Updated
Jul 31, 2026 - Python
Add this topic to your repo
To associate your repository with the benchmark-tasks topic, visit your repo's landing page and select "manage topics."