{"id":51276566,"url":"https://github.com/primer/agent-eval","last_synced_at":"2026-06-29T21:03:13.552Z","repository":{"id":358118855,"uuid":"1240100927","full_name":"primer/agent-eval","owner":"primer","description":"Tools to evaluate agent performance on Primer tasks","archived":false,"fork":false,"pushed_at":"2026-06-11T16:45:30.000Z","size":209,"stargazers_count":0,"open_issues_count":4,"forks_count":0,"subscribers_count":0,"default_branch":"main","last_synced_at":"2026-06-11T18:08:42.681Z","etag":null,"topics":[],"latest_commit_sha":null,"homepage":"","language":"TypeScript","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"mit","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/primer.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":".github/CODE_OF_CONDUCT.md","threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":".github/SECURITY.md","support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null,"notice":null,"maintainers":null,"copyright":null,"agents":"AGENTS.md","dco":null,"cla":null}},"created_at":"2026-05-15T19:11:58.000Z","updated_at":"2026-06-11T16:45:43.000Z","dependencies_parsed_at":null,"dependency_job_id":null,"html_url":"https://github.com/primer/agent-eval","commit_stats":null,"previous_names":["primer/agent-eval"],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/primer/agent-eval","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/primer%2Fagent-eval","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/primer%2Fagent-eval/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/primer%2Fagent-eval/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/primer%2Fagent-eval/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/primer","download_url":"https://codeload.github.com/primer/agent-eval/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/primer%2Fagent-eval/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":34942665,"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":"2026-06-29T02:00:05.398Z","response_time":58,"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":"2026-06-29T21:03:08.990Z","updated_at":"2026-06-29T21:03:13.546Z","avatar_url":"https://github.com/primer.png","language":"TypeScript","funding_links":[],"categories":[],"sub_categories":[],"readme":"# agent-eval\n\n\u003e Tools for evaluating the performance of agents on Primer-related tasks.\n\n## Overview\n\nThis project creates a framework used for running experiments. In each\nexperiment, we establish treatments that are then used to setup the environment\nfor the agent before completing evaluations. These evaluations (evals) represent\na scenario where, given a prompt, we are evaluating how the agent behaves.\n\nThis framework allows us to test the effectiveness of different experimental\ntreatments on the evaluations we care about. In particular, we score results\nbased on:\n\n- Correctness: how many of the tests does the agent output pass\n- Cost: how much did the agent spend in API calls\n- Latency: how long did the agent take to complete the evals\n\n## Experiments\n\nExperiments live in [`./experiments/src`](./experiments/src/). Each experiment\nis a new TypeScript file that exports an experiment config:\n\n```ts\nimport type {ExperimentConfig} from '@primer/agent-experiment'\n\nexport const experiment: ExperimentConfig = {\n  name: 'Example experiment',\n  description: 'Experiment config demonstrating different options',\n  models: ['gpt-5.5', 'claude-opus-4.7', 'claude-sonnet-4.6'],\n  evals: ['001-agent-uses-button-from-primer', '002-agent-uses-octicon-from-primer'],\n  treatments: [\n    {\n      name: 'Treatment one',\n      async setup({sandbox}) {\n        // ...\n      },\n    },\n    {\n      name: 'Treatment two',\n      async setup({sandbox}) {\n        // ...\n      },\n    },\n  ],\n}\n```\n\nThe experiment config will specify:\n\n- A name for the experiment\n- A description for the experiment describing more about what it is for\n- Models against which you would like to run the experiment\n- Evaluations (evals) that are used to grade the output of the model\n- Treatments that specify the different conditions you would like to test (for\n  example, testing with an MCP server versus without)\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fprimer%2Fagent-eval","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fprimer%2Fagent-eval","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fprimer%2Fagent-eval/lists"}