recruiting-copilot: An AI Recruiting Workflow Plugin for DeepSeek Harness
viy1204/recruiting-copilot
A DSH plugin delivering an AI recruiting workflow for HR/headhunters: role profiling, dual-channel sourcing, resume screening, interview scheduling, tracking and daily reports, with an operable recruiting browser panel.
recruiting-copilot is an AI recruiting workflow plugin for DeepSeek Harness, designed for HR and headhunters to automate the entire chain from job clarification, dual-platform sourcing and screening, market mapping, resume evaluation, interview scheduling, to candidate ledger and daily reports. It turns vague requirements into executable criteria through one-question-at-a-time interviews, and a daily 'handle today's recruiting' command automatically checks unread messages, searches candidates, screens, and generates reports. The built-in 'recruiting browser' panel offers intuitive operation while defaulting to read-only with safety confirmations to protect accounts. With it, daily recruiting shifts from tedious manual work to a conversational automated flow.
How to Install
dsh plugin --profile web add github:viy1204/recruiting-copilot- Category
- Workflow & Automation
- Platform
- DSH Plugin
- Author
- viy1204
- Distribution
- Plugin
recruiting-copilot Key Features
recruiting-copilot Repo Summary
What Does It Do?
recruiting-copilot is a DSH plugin for DeepSeek Harness that turns the daily recruiting workflow for HR and headhunters into an automated pipeline. It solves the pain of vague job requirements, time-consuming sourcing and screening, tedious resume evaluation, and coordination-heavy interview scheduling, letting you handle everything from job clarification to candidate tracking and daily reports directly inside the harness. Core capabilities include job requirement clarification, dual-channel sourcing on Boss Zhipin and Liepin, market talent mapping, resume evaluation, interview scheduling, and candidate ledger with daily reports, plus a built-in "recruiting browser" panel for direct operation.
Core Features
- Job Clarification: turns vague hiring needs into executable screening criteria through one-question-at-a-time interviews, producing both external JD and internal sourcing notes.
- Daily Recruiting: say "handle today's recruiting" and it automatically checks unread messages, searches candidates, screens them, greets after your confirmation, updates the ledger, and generates a daily report that can be pushed as a Feishu card.
- Market Mapping: deep-dives into market supply, salary levels, and talent availability at target companies, with an actionable list of candidates.
- Resume Intake & Evaluation: supports local resumes or fetching Liepin/BOSS resume attachments from Feishu email, deduplicating and running them through the same review and ledger flow.
- Interview Scheduling: checks interviewer availability, creates calendar events with video conference links, invites interviewers, and generates candidate invitation scripts, with interview profiles auto-synced.
- Improves Over Time: extracts search terms from real matched resumes each round and feeds them back into the keyword iteration table for more accurate future searches.
How to Use This Plugin?
After installing and restarting the DSH session, the 7 skills under the repo's skills/ directory become available in any workspace's skill directory. On first use, run the init skill to set up a workspace, then say "help me initialize the recruiting workspace" for any role to start clarifying requirements; for daily use, open your workspace and say "handle today's recruiting." When unsure, ask "how do I use these tools" and the AI will guide you through the master index. To adjust behavior, modify the keyword iteration table or the safety confirmation toggle in the configuration.
This page is an independent rewrite of the plugin's official README — for authoritative documentation and the latest changes, refer to the source: viy1204/recruiting-copilot. The plugin is third-party code that runs on your machine once installed; inclusion does not imply endorsement — please review the source before installing.