{"id":31655614,"url":"https://github.com/przeprogramowani/prd-eval","last_synced_at":"2025-10-07T13:17:42.811Z","repository":{"id":317750985,"uuid":"1068690553","full_name":"przeprogramowani/prd-eval","owner":"przeprogramowani","description":"PRD Evals","archived":false,"fork":false,"pushed_at":"2025-10-02T20:59:25.000Z","size":190,"stargazers_count":0,"open_issues_count":0,"forks_count":0,"subscribers_count":0,"default_branch":"master","last_synced_at":"2025-10-02T21:13:27.372Z","etag":null,"topics":[],"latest_commit_sha":null,"homepage":null,"language":"TypeScript","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/przeprogramowani.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"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":"AGENTS.md","dco":null,"cla":null}},"created_at":"2025-10-02T19:01:23.000Z","updated_at":"2025-10-02T20:59:28.000Z","dependencies_parsed_at":"2025-10-02T21:14:37.477Z","dependency_job_id":"e46fb01a-b249-454c-af61-18e55e282700","html_url":"https://github.com/przeprogramowani/prd-eval","commit_stats":null,"previous_names":["przeprogramowani/prd-eval"],"tags_count":null,"template":false,"template_full_name":null,"purl":"pkg:github/przeprogramowani/prd-eval","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/przeprogramowani%2Fprd-eval","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/przeprogramowani%2Fprd-eval/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/przeprogramowani%2Fprd-eval/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/przeprogramowani%2Fprd-eval/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/przeprogramowani","download_url":"https://codeload.github.com/przeprogramowani/prd-eval/tar.gz/refs/heads/master","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/przeprogramowani%2Fprd-eval/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":278780261,"owners_count":26044526,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","status":"online","status_checked_at":"2025-10-07T02:00:06.786Z","response_time":59,"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":"2025-10-07T13:17:37.488Z","updated_at":"2025-10-07T13:17:42.804Z","avatar_url":"https://github.com/przeprogramowani.png","language":"TypeScript","funding_links":[],"categories":[],"sub_categories":[],"readme":"# PRD evaluation with promptfoo\n\nThis repository compares product requirement documents generated by different model vendors. The TypeScript tooling under `src/` prepares a promptfoo run, executes three judges (GPT-5, Claude Sonnet 4.5, and Gemini 2.5 Pro), and aggregates their JSON responses into repeatable score reports.\n\n## Prerequisites\n\n- Node.js 18 or newer\n- API access for every judge configured in `src/promptfoo.config.ts` (set in `.env`):\n  - `OPENAI_API_KEY`\n  - `ANTHROPIC_API_KEY`\n  - `OPENROUTER_API_KEY`\n\n## Installation\n\n```bash\nnpm install\n```\n\n## Running evaluations\n\n```bash\nnpm run eval\n```\n\nThe evaluation script clears `dist/results`, rebuilds the TypeScript sources, disables the promptfoo cache, and runs `dist/evalRunner.js`. Each judge receives the rubric in `prd-job-interview.md` plus the vendor PRDs listed in `src/promptfoo.config.ts`. Outputs are collected in:\n\n- `dist/results/latest.json` – raw promptfoo output\n- `dist/results/latest.lite.json` – condensed per-judge scores without prompts\n- `dist/results/score-history.csv` – append-only log of every run and criterion score\n\n## Inspecting results locally\n\n```bash\nnpm run display-scores\n```\n\nThis command re-parses `latest.lite.json` and prints a Markdown summary table (or writes to the GitHub Actions step summary when running in CI).\n\n## Data layout\n\n- `data/{vendor}/prd.md` contains the full PRD for a model run.\n- `data/{vendor}/summary.md` captures the at-a-glance synopsis of the same run.\n- Vendor folder names stay in lowercase kebab case (for example `data/sonnet-45/`).\n\nDuplicate an existing vendor folder when starting a new run, prune unused drafts with `rg --files data`, and update `modelDirectories` in `src/promptfoo.config.ts` so evaluators pick up the new PRD.\n\n## Configuration overview\n\n- Prompt template and judge list: `src/promptfoo.config.ts`\n- Evaluation orchestrator and result logging: `src/evalRunner.ts`\n- Markdown score summariser: `src/resultInterpreter.ts`\n- Rubric applied to every run: `prd-job-interview.md`\n\nThe prompt template enforces strict JSON output, assigns judge aliases, and instructs models to score five criteria plus an overall verdict. `evalRunner` normalises the JSON (stripping code fences or \"thinking\" blocks) before recording results in CSV and lite JSON formats.\n\n## Maintenance tips\n\n- Run `npm run build` (and optionally `npm run eval`) before committing to ensure the generated JavaScript is up to date.\n- Lint documentation with `npx markdownlint-cli2 \"data/**/*.md\"` and audit for placeholders using `rg \"TODO\" data`.\n- Never edit `dist/` by hand—rebuild instead so generated artifacts stay traceable.\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fprzeprogramowani%2Fprd-eval","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fprzeprogramowani%2Fprd-eval","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fprzeprogramowani%2Fprd-eval/lists"}