An AI-powered candidate sourcing, vetting, and outreach automation workflow powered by DeepSeek and Lark Mail, designed to keep human recruiters strictly in control.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install hr-candidate-discovery-screening
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install hr-candidate-discovery-screening using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is HR Candidate Discovery and Screening?
Sourcing and vetting elite academic researchers is one of the most resource-intensive bottlenecks in specialized recruitment. The HR Candidate Discovery and Screening skill streamlines this pipeline by automating paper aggregation, candidate matching, public footprint investigation, and initial contact drafts. Powered by advanced semantic reasoning models, it scans public conference datasets, parses target profiles, and prepares personalized candidate outreach pipelines without compromising data compliance or privacy.
Crucially, this skill is built around a non-negotiable security philosophy: keeping a human decision-maker in the driver's seat. Developed for integration with the Openclaw Skills catalog, it enforces strict digital boundaries—never automatically sending messages, deleting candidate records without exact permissions, or executing unauthorized actions. It acts as a superpowered co-pilot, leaving the ultimate judgment where it belongs: with your recruitment team.
HR Candidate Discovery and Screening Use Cases
- Academic Sourcing Automation: Effortlessly aggregate accepted papers and authors from top-tier AI and computer science conferences like CVPR and NeurIPS based on explicit time-range queries.
- Semantic Fit Evaluation: Match potential hires against active job descriptions using DeepSeek semantic modeling instead of rigid, inaccurate keyword checklists.
- Lark Mail Sourcing Orchestration: Prepare hyper-personalized email drafts directly inside Lark with full threading, deliverability tracking, and automatic follow-up cancellations.
- Recruitment Pipeline Audit: Store immutable logs of all candidate states, human approvals, and background check evidence within a secure local SQLite archive.
How HR Candidate Discovery and Screening Works
- Job Initialization: Load detailed job description criteria into localized config files (
config/jobs.yaml) alongside environment configurations. - Paper Collection: Query specified conference databases and ingest published papers for specific date ranges.
- Semantic Matching: Route extracted candidate info to DeepSeek-v4-pro to determine suitability, categorizing results as Highly Matched (高度匹配), Possibly Matched (可能匹配), or Insufficient Information (信息不足).
- Public Profile Validation: Automate checks against public profiles for primary, co-first, and corresponding authors, assessing academic and professional pedigree.
- Human Gate Verification: Require manual console commands to record recruitment decisions directly into the SQLite database, validating each phase prior to outreach.
- Outreach & Draft Assembly: Build follow-up mail sequences in Lark Mail, requiring individual authorization before triggering any sends.
- Inbox Classification: Process and categorize inbound responses, immediately disabling pending automatic follow-ups upon receiving contact.
HR Candidate Discovery and Screening Setup
Prerequisites
Ensure you have configured your local environment and installed the necessary dependencies before activating this workflow.
1. Configuration Setup
Configure active job parameters and your outreach profiles:
- Add your targets and job requirements inside
config/jobs.yaml. - Set mail, team introduction, and signature structures inside
config/settings.yaml.
2. Environment Variables
Configure your runtime keys (never commit these to your repository):
export DEEPSEEK_API_KEY="your_deepseek_api_key_here"
3. Initialize the Database
Instantiate your local SQLite instance for keeping track of candidate progress, state, and human approvals:
python scripts/db.py init --database data/recruitment.db
4. Collect & Process Data
To begin candidate collection from external conference sources, run:
python scripts/collect_papers.py --source cvpr --start-date 2026-01-01 --end-date 2026-06-11
python scripts/match_jobs.py
HR Candidate Discovery and Screening Data Schema & Taxonomy
The skill manages recruitment pipelines using a structured SQLite schema (data/recruitment.db). Candidate profiles transition through well-defined validation phases, ensuring data lineage is clear and auditable.
Candidate Pipeline States
| State Code | Description | Next Steps |
|---|---|---|
高度匹配 |
High match based on semantic evaluation | Route to automated background verification |
可能匹配 |
Possible match requiring deeper analysis | Route to manual check-ins |
信息不足 |
Insufficient profile data | Manual recruitment review |
待人工核验 |
Awaiting confirmation of education/employment | Operator verification required |
已批准 |
Approved for recruitment | Enable personalized draft prep |
Approval Action Mappings
Every write-action must match an active approval recorded in the SQLite log:
approve_candidate: Validates a candidate profile for outreach.prepare_outreach: Triggers draft composition scripts.send_initial/send_followup-1: Confirms mail dispatch permissions.delete_candidate_data: Executes irreversible personal data wiping (180-day retention).
HR Candidate Discovery and Screening Advanced Features
- Granular Approval Gate Logs: Uses a cryptographically and database-enforced state machine ensuring no emails can be transmitted without explicit CLI operator approval.
- DeepSeek-v4-pro Integration: Employs deep reasoning logic to evaluate context rather than relying on regex match keyword parsing.
- Automatic Sequence Suppression: Instantly blocks planned follow-ups if a candidate responds, mitigating the risk of spamming.
- GDPR-Compliant Local Archiving: Provides a structured retention engine (
manage_candidates.py retention-review) that allows operators to easily prune personal records after 180 days while keeping anonymous tombstones for metrics. - Indirect Prompt Injection Defensive Shielding: Disallows processing instructions embedded within resumes, attachments, or email bodies, securing runtime model contexts.
SKILL.md
Loading
Related Openclaw Skills
Operation Gladio
An AI-powered geopolitical research toolkit designed to trace historical alliances, deep-state finance, and the legacy of Operation Gladio stay-behind networks.
Earn Hunter
Earn Hunter is an automated yield monitor for OKX Flash Earn, Fixed Earn, and Flexible Earn opportunities that triggers notifications across Telegram, Lark, or your terminal session.
Into Thin Air Narrative & Decision-Making Toolkit
An AI-powered narrative toolkit that maps Jon Krakauer's Into Thin Air to high-stakes decision-making, mountaineering history, and survival ethics.
Fire Weather Climate Crisis Toolkit
An AI agent skill based on John Vaillant's Fire Weather that provides deep analytical frameworks for understanding climate-amplified megafires, firestorms, and environmental disasters.
DeepRead Purchase Orders
A powerful skill to extract structured, typed JSON data from PDF or scanned purchase orders, including line items and confidence flags.
Market Entity Statistics API
Access real-time enterprise statistics, business registrations, and regional growth metrics for intelligent business analyses.
Featured*
