{"id":"891e40b5-cad8-4698-b7e3-9528563ba277","slug":"hr-toolkit","kind":"skill","name":"HR-toolkit","summary":"Composable Codex plugin for conference papers, domain matching, public person enrichment, and Excel delivery.","owner":"SeanX16","repoUrl":"https://github.com/SeanX16/HR-toolkit","license":"MIT","licenseTier":"permissive","trust":"community","stars":0,"match":null,"fewStars":true,"format":"skill","createdAt":"2026-07-31T06:44:27.000Z","addedAt":"2026-10-21T12:35:57.339Z","logoUrl":"https://avatars.githubusercontent.com/u/69029679?u=509d51fcefdf06cb81520603706677a21f28c648&v=4","linkOnly":false,"collection":null,"skillPath":null,"categories":["files-documents"],"tags":["codex","codex-plugin","conference-papers","excel","hr-tech","openai","talent-research"],"runtime":"binary","local":true,"cloudCapable":false,"cloudCapability":"unknown","cloudReadyOnPlatform":false,"needsApiKey":false,"reviewed":false,"syncedAt":"2026-10-23T05:19:02.245Z","sample":false,"url":"https://conduits.dev/tools/hr-toolkit","language":"Python","transports":["stdio"],"install":{},"requirements":{"env":[],"upstreamMayCharge":false},"tools":null,"readme":"# HR Toolkit\n\n[中文说明](README_zh.md)\n\nHR Toolkit is a composable, skills-only plugin for Codex. It installs four\nindependent Skills together and lets Codex select one Skill or combine several\naccording to the task.\n\nIt supports two related workflows:\n\n- build a complete, source-backed conference paper and author roster;\n- identify papers related to a research domain, enrich public professional\n  information, and deliver a review-ready Excel workbook.\n\nIt can also start from a general list of people, without requiring conference\nor paper inputs.\n\n## Included Skills\n\n| Stage | Skill | Responsibility | Main output |\n| --- | --- | --- | --- |\n| A | `collect-conference-papers` | Collect papers and author appearances from declared conference sources | `conference-paper-catalog/v1` |\n| B | `match-papers-to-domain` | Turn keywords into weighted concepts and match every paper with evidence | `paper-domain-matches/v1` |\n| C | `enrich-person-records` | Confirm a field preset or custom definitions, then enrich public professional information | `author-enrichment/v1` or `person-enrichment/v1` |\n| D | `render-talent-tables` | Turn structured inputs into a standard or natural-language-customized Excel workbook | `.xlsx` |\n\nCommon routes:\n\n```text\nConference roster:       A\nDomain-focused research: A -> B -> C\nConference Excel:        A -> B -> C -> D\nGeneral people list:     C\nCustom people Excel:     C -> D\nExisting JSON to Excel:  D\n```\n\nThere is no fifth routing Skill. Each Skill has distinct activation metadata,\nand Codex decides which one or which combination applies.\n\nB does not weight every supplied phrase mechanically. It can group aliases into\none concept, retain conference-wide umbrella terms as non-scoring context, and\navoid forcing newer labels onto older papers.\n\nBefore C researches any person, it presents Preset A (professional core),\nPreset B (academic talent full), and Custom. Custom fields require explicit\ndefinitions, formats, evidence rules, ambiguity handling, and user confirmation.\nAfter a custom run, the user can choose whether to save the field set as a preset.\n\n## Install from GitHub\n\nAdd this repository as a Codex plugin marketplace, then install the bundled\nplugin:\n\n```powershell\ncodex plugin marketplace add SeanX16/HR-toolkit --ref main\ncodex plugin add hr-toolkit@hr-toolkit\n```\n\nAfter installation, start a new Codex task and try one of these prompts:\n\n```text\nCollect all papers from this conference and preserve the declared source scope.\nMatch these papers to neural rendering and show the evidence for every paper.\nEnrich this CSV of names with current organization, role, and public profile.\nTurn this JSON into a review-ready Excel workbook with the columns I describe.\n```\n\nThe repository marketplace entry is stored at\n`.agents/plugins/marketplace.json`. The installable package under\n`plugins/hr-toolkit/` is generated from the source and should not be edited\nmanually.\n\nFor current plugin and marketplace concepts, see the\n[official Codex plugin packaging documentation](https://developers.openai.com/plugins/build/plugins).\n\n## Inputs and contracts\n\nThe conference workflow uses:\n\n```text\nconference-paper-catalog/v1\n        |\n        v\npaper-domain-matches/v1\n        |\n        v\nauthor-research-result/v1\n        |\n        v\nauthor-enrichment/v1\n```\n\nThe general people-list workflow uses:\n\n```text\npasted names / CSV / JSON / profile URLs\n        |\n        v\nperson-seed-list/v1\n        |\n        v\nperson-field-selection/v1 (user-confirmed)\n        |\n        v\nperson-research-result/v1\n        |\n        v\nperson-enrichment/v1\n```\n\nAll contracts preserve stable IDs, upstream fingerprints, collection time,\nsource URLs, unknown values, and review state.\n\n## Safety and intended use\n\nHR Toolkit is a research and delivery workflow, not a hiring decision system.\n\n- It uses public, person-specific professional evidence.\n- Missing or ambiguous information stays `Unknown` or `Needs Review`.\n- It does not infer nationality, ethnicity, age, gender, religion, disability,\n  marital status, political affiliation, or other protected attributes from a\n  name.\n- Domain matching describes paper-topic evidence; it is not a candidate score,\n  hiring recommendation, rejection, or outreach decision.\n- Local runs, researched people data, email lists, and generated workbooks are\n  excluded from Git.\n\nRead [PRIVACY.md](PRIVACY.md) before using real person data and\n[SECURITY.md](SECURITY.md) before reporting a vulnerability.\n\n## Repository layout\n\n```text\n.codex-plugin/   Plugin manifest\n.agents/         Repository marketplace metadata\nskills/          The only editable source for the four Skills\npackages/        Shared deterministic Python runtime\nschemas/         Versioned cross-Skill JSON schemas\nplugins/         Generated installable plugin package committed for distribution\ntests/           Offline contract and integration tests\ntools/           Validation and packaging utilities\n.work/           Local run data, ignored by Git\noutputs/         Local Excel deliveries, ignored by Git\ndistributions/   Optional local build output, ignored by Git\n```\n\n## Development\n\nRequirements:\n\n- Python 3.11 or newer\n- Node.js for the D-stage renderer syntax check\n- A Codex workspace runtime with `@oai/artifact-tool` when creating Excel files\n\nRun the source validation:\n\n```powershell\npython tools/validate_project.py\n```\n\nRebuild and verify the repository marketplace package:\n\n```powershell\npython tools/build_distributions.py --output plugins --overwrite\npython tools/verify_distributions.py --root plugins\n```\n\nBuild an optional standalone local distribution:\n\n```powershell\npython tools/build_distributions.py --overwrite\npython tools/verify_distributions.py\n```\n\nSee [CONTRIBUTING.md](CONTRIBUTING.md) for change and pull-request guidance.\n\n## License\n\n[MIT](LICENSE)\n"}