{"id":50900089,"url":"https://github.com/teaql/teaql-agent-kit","last_synced_at":"2026-06-24T00:02:06.969Z","repository":{"id":37522385,"uuid":"162171043","full_name":"teaql/teaql-agent-kit","owner":"teaql","description":"Deterministic execution for non-deterministic AI.","archived":false,"fork":false,"pushed_at":"2026-06-19T00:04:18.000Z","size":39784,"stargazers_count":2799,"open_issues_count":0,"forks_count":955,"subscribers_count":100,"default_branch":"main","last_synced_at":"2026-06-19T02:06:14.563Z","etag":null,"topics":["agentic-coding","ai-coding","code-generation","coding-agent","ddd","developer-tools","domain-driven-design","java","rust","semantic-guardrails","semantic-modeling","teaql"],"latest_commit_sha":null,"homepage":"https://teaql.io","language":"Python","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/teaql.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","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":"AGENTS.md","dco":null,"cla":null}},"created_at":"2018-12-17T18:07:12.000Z","updated_at":"2026-06-18T08:09:41.000Z","dependencies_parsed_at":"2024-06-25T13:24:54.867Z","dependency_job_id":null,"html_url":"https://github.com/teaql/teaql-agent-kit","commit_stats":null,"previous_names":["doublechaintech/retailscm-biz-suite","teaql/teaql-agent-kit"],"tags_count":1,"template":false,"template_full_name":null,"purl":"pkg:github/teaql/teaql-agent-kit","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/teaql%2Fteaql-agent-kit","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/teaql%2Fteaql-agent-kit/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/teaql%2Fteaql-agent-kit/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/teaql%2Fteaql-agent-kit/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/teaql","download_url":"https://codeload.github.com/teaql/teaql-agent-kit/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/teaql%2Fteaql-agent-kit/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":34711203,"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":"2026-06-23T02:00:07.161Z","response_time":65,"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":["agentic-coding","ai-coding","code-generation","coding-agent","ddd","developer-tools","domain-driven-design","java","rust","semantic-guardrails","semantic-modeling","teaql"],"created_at":"2026-06-16T02:00:25.215Z","updated_at":"2026-06-24T00:02:06.957Z","avatar_url":"https://github.com/teaql.png","language":"Python","funding_links":[],"categories":["miscellaneous","Catalog"],"sub_categories":["Evaluation Harnesses \u0026 Benchmarks"],"readme":"# TEAQL Agent Kit\n\n**TEAQL Agent Kit** is an evaluation environment for coding agents and language models on auditable business software tasks.\n\nIt is designed to measure not only whether generated code works, but also whether an agent can preserve business semantics, follow framework boundaries, maintain auditability, recover from errors, and use tokens efficiently.\n\n---\n\n## What This Repository Is\n\nThis repository is focused on **evaluation**.\n\nIt provides tasks, prompts, guides, and reports for observing how coding agents behave when working with TEAQL-based business software.\n\nThe current goal is not ungated production automation.\n\nThe current goal is to answer a more basic question:\n\n\u003e How do coding agents actually behave when business rules, generated APIs, audit traces, and framework boundaries matter?\n\n---\n\n## Main Branch: Controlled Evaluation\n\nThe `main` branch is the primary entry point.\n\nIt is used for controlled and reproducible evaluation, with:\n\n* Clear task definitions\n* Explicit TEAQL API rules\n* Agent-readable guides\n* Optional human checkpoints\n* Comparable evaluation reports\n\nThis branch asks:\n\n\u003e What can a coding agent do when the rules are clear and the evaluation is controlled?\n\n---\n\n## Autonomous Branch: No-Gate Evaluation\n\nThe `autonomous` branch is for experimental no-gate evaluation.\n\nIt is used to observe how far coding agents can go without human intervention checkpoints.\n\nThis branch focuses on:\n\n* Fully automatic task attempts\n* Self-repair behavior\n* Unsafe shortcuts\n* Framework boundary violations\n* Token usage\n* Guardrails that may be needed before production use\n\nThe autonomous branch is for benchmarking and stress-testing.\nIt is not a recommendation for ungated production deployment.\n\nThis branch asks:\n\n\u003e What does a coding agent actually do when no human gate is present?\n\n---\n\n## Evaluation Focus\n\nTEAQL Agent Kit evaluates agents across dimensions such as:\n\n* Functional completion\n* API adherence\n* Hallucinated API count\n* Audit coverage\n* Framework discipline\n* Error recoverability\n* Human intervention count\n* Token efficiency\n\nFor long-lived business software, these dimensions matter as much as whether the code compiles.\n\n---\n\n## Reports\n\nEvaluation reports will be published in this repository.\n\nReports may include controlled and autonomous runs across different coding agents, language models, and TEAQL stacks.\n\n---\n\n## Evaluation Across Stacks\n\nTEAQL Agent Kit may evaluate equivalent business software tasks across different TEAQL implementations, including:\n\n- TEAQL Java stack\n- TEAQL Rust stack\n\nThe purpose is not to rank programming languages.\n\nThe purpose is to understand how coding agents preserve semantics, auditability, and framework boundaries across different implementation stacks.\n\n---\n\n## Long-Term Direction\n\nToday, TEAQL Agent Kit evaluates coding agents.\n\nLong term, the same evidence may help define which AI coding tasks can be safely automated, which require human gates, and which should never bypass review.\n\nThe goal is measured automation, not blind automation.\n\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fteaql%2Fteaql-agent-kit","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fteaql%2Fteaql-agent-kit","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fteaql%2Fteaql-agent-kit/lists"}