{"id":51220950,"url":"https://github.com/harbor-framework/benchmark-template","last_synced_at":"2026-06-28T07:03:05.834Z","repository":{"id":334139931,"uuid":"1140191077","full_name":"harbor-framework/benchmark-template","owner":"harbor-framework","description":"Harbor Benchmark Template","archived":false,"fork":false,"pushed_at":"2026-06-25T00:06:49.000Z","size":1091,"stargazers_count":13,"open_issues_count":15,"forks_count":10,"subscribers_count":0,"default_branch":"main","last_synced_at":"2026-06-25T02:05:25.952Z","etag":null,"topics":[],"latest_commit_sha":null,"homepage":"https://harborframework.com/","language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":null,"status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/harbor-framework.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":"CONTRIBUTING.md","funding":null,"license":null,"code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null,"notice":null,"maintainers":null,"copyright":null,"agents":null,"dco":null,"cla":null}},"created_at":"2026-01-23T00:14:55.000Z","updated_at":"2026-06-25T00:06:52.000Z","dependencies_parsed_at":null,"dependency_job_id":null,"html_url":"https://github.com/harbor-framework/benchmark-template","commit_stats":null,"previous_names":["harbor-framework/benchmark-template","harbor-framework/benchmark"],"tags_count":0,"template":true,"template_full_name":null,"purl":"pkg:github/harbor-framework/benchmark-template","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/harbor-framework%2Fbenchmark-template","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/harbor-framework%2Fbenchmark-template/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/harbor-framework%2Fbenchmark-template/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/harbor-framework%2Fbenchmark-template/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/harbor-framework","download_url":"https://codeload.github.com/harbor-framework/benchmark-template/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/harbor-framework%2Fbenchmark-template/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":34880191,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-05-26T15:22:16.424Z","status":"online","status_checked_at":"2026-06-28T02:00:05.809Z","response_time":54,"last_error":null,"robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":true,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":[],"created_at":"2026-06-28T07:03:04.813Z","updated_at":"2026-06-28T07:03:05.828Z","avatar_url":"https://github.com/harbor-framework.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"# Benchmark Template\n\nThis is a template for building agent benchmarks with [Harbor](https://harborframework.com).\n\nIt provides structure, CI workflows, and example tasks to help you build your own benchmark following [best practices](https://www.anthropic.com/engineering/demystifying-evals-for-ai-agents) and leveraging automation where it is appropriate. Everything is designed to be customized to your benchmark.\n\n## What's Included\n\n- **[Contributing Guide](CONTRIBUTING.md)**: guide for creating, testing, and submitting tasks\n  - [PR Checklist](.github/pull_request_template.md): submission template for contributors\n- **[Task Proposal Rubric](rubrics/task-proposal.md)**: LLM evaluation criteria for task proposals (Discussions, Discord)\n  - [Discord Review Bot](tools/discord-review-bot/): auto-reviews proposals in Discord\n  - [Discussion Review Workflow](.github/workflows/discussion-review.yml): auto-reviews proposals in GitHub Discussions\n  - [Batch Grader](tools/batch-grader/): grade all tasks against the rubric, useful for iterating on it\n- **[Task Review Automation](TASK_REVIEW_AUTOMATION.md)**: CI automation for validating task PRs\n  - [Static checks](TASK_REVIEW_AUTOMATION.md#static-checks): path validation, Dockerfile sanity, canary, metadata, test references\n  - [Implementation rubric review](TASK_REVIEW_AUTOMATION.md#implementation-rubric-review): `harbor check` with [27-criteria rubric](rubrics/task-implementation.toml) (auto on PR)\n  - [Similarity check](TASK_REVIEW_AUTOMATION.md#check-similarity): TF-IDF duplicate detection\n  - [Docker build + oracle/nop validation](TASK_REVIEW_AUTOMATION.md#docker-build-oracle-nop): environment builds, solution passes, no-op fails\n  - [Agent trials](TASK_REVIEW_AUTOMATION.md#agent-trials): multi-agent runs with analysis\n  - [Cheat trials](TASK_REVIEW_AUTOMATION.md#cheat-trials): adversarial reward-hack detection\n- **[Test Tasks](ci_checks/test-tasks/)**: intentionally broken tasks for regression testing the QA pipeline\n- **[Reviewing Guide](REVIEWING.md)**: guide for human reviewers evaluating task PRs\n\n## Getting Started\n\n#### 1. Create your repo\n\nClick \"Use this template\" on GitHub.\n\n#### 2. Customize\n\nGrep for `CUSTOMIZE` in source files to find what to edit.\n\n\u003e [!TIP]\n\u003e Edit [`rubrics/task-implementation.toml`](rubrics/task-implementation.toml) to define criteria for evaluating task implementations on PRs.\n\n\u003e [!TIP]\n\u003e Edit [`rubrics/task-proposal.md`](rubrics/task-proposal.md) to define criteria for evaluating task proposals on Discussions and Discord.\n\n\u003e [!TIP]\n\u003e Edit [`.github/harbor-run-defaults.yml`](.github/harbor-run-defaults.yml) to change default agents, trial count, timeout, and trigger.\n\n#### 3. Set repo secrets\n\n| Secret | Used by | Required? |\n|--------|---------|-----------|\n| `ANTHROPIC_API_KEY` | Implementation rubric review | Yes |\n| `GPTZERO_API_KEY` | AI detection (optional, not in default workflow) | Optional |\n| `OPENAI_API_KEY` | `/run` trials | Optional |\n| `GEMINI_API_KEY` | `/run` trials | Optional |\n| `MODAL_TOKEN_ID`, `MODAL_TOKEN_SECRET` | `/run`, `/cheat`, and `/validate` when `env: modal` / `validate_env: modal` is set in `harbor-run-defaults.yml` (or comment override) | Optional |\n\n#### 4. Get future improvements\n\n```bash\n# One-time setup\ngit remote add template https://github.com/harbor-framework/benchmark-template.git\n\n# Pull latest\ngit fetch template\ngit merge template/main\n```\n\n#### 5. Join the community\n\nJoin our active [Terminal Bench community](https://discord.gg/ZvcWupVXjz) to share your benchmark or get help.\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fharbor-framework%2Fbenchmark-template","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fharbor-framework%2Fbenchmark-template","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fharbor-framework%2Fbenchmark-template/lists"}