recruitment-workflow
Skill0 starsChanged 38 days agoMITView on GitHub
Coordinate a recruitment assignment from intake to candidate presentation, select the appropriate recruiter skill, and track evidence, owners, approvals, and next actions.
In this repository11 other files it gives its agents.
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- candidate-outreachSkill
- candidate-presentationSkill
- candidate-screeningSkill
- hiring-scorecardSkill
- interview-invitationSkill
- interview-kitSkill
- job-intakeSkill
- sourcing-strategySkill
- vitaeSkill
The file
We split this skill into the rules it gives an agent
1 rule. About 0 are common, 1 is distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈1
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 40
Do not auto-reject or make employment decisions.
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Recruitment workflow
Establish the assignment
Identify the role, client, workspace if connected, hiring owner, existing progress, and desired outcome. Reuse the actual current stage rather than restarting intake. Read only the records required for the assignment. Do not turn a request for one email into a full recruiting project.
Route the next step
| Situation | Skill | Required handoff |
| Requirements are unclear | job-intake | Brief and hiring-owner decisions |
| Evaluation criteria are missing | hiring-scorecard | Agreed criteria, evidence, and anchors |
| Need relevant candidates | sourcing-strategy | Search plan, approved channels/spend, provenance |
| Have profiles to assess | candidate-screening | Evidence report and recruiter decision |
| Need to approach a candidate | candidate-outreach | Verified facts, draft, recipient history |
| Interview is the agreed next step | interview-invitation | Confirmed/proposed logistics and booking state |
| Interviewers need a plan | interview-kit | Timed questions and assessment guidance |
| Client needs a candidate summary | candidate-presentation | Evidence and approved sharing scope |
Load only the relevant installed skill. Skills may be installed individually; never assume a sibling directory is present. If a needed skill is missing, explain its purpose and use bunx skills add vitaedotai/agent --skill <name> only when installation is within the user's request. Otherwise give a concise manual handoff. Do not claim that delegation or installation happened when it did not.
Maintain state
Keep a compact record with: assignment ID or descriptive label, current stage, artifact/version, evidence source, responsible person, unresolved questions, approval state, and next action with due date if supplied. Use actual Vitae records/tasks when verified tools support the requested persistence. Otherwise return a handoff table for the recruiter to save; do not claim chat memory is durable project state.
Possible states are draft, ready for review, awaiting user decision, authorized, queued, completed, and failed. They are not interchangeable. Before resuming a send, record creation, import, or workflow after a timeout, inspect the current state to avoid duplication.
Execution boundaries
An agreed job brief does not authorize paid searches, candidate outreach, rejection, or public sharing. Honor explicit authorization already supplied for the same action and scope, and preserve Vitae's server approval gates. Do not repeatedly ask approval for ordinary reads or writing a requested draft.
Do not auto-reject or make employment decisions. Use job-related evidence and human review. Keep candidate information out of public artifacts. If a connector lacks a needed action, produce the artifact and identify the exact manual step; do not substitute another tool with broader effects.
Beyond the initial catalog
The first release covers intake through candidate presentation. Offers, reference checks, rejection communication, placement handover, and post-placement follow-up remain explicit recruiter-owned steps. Record their owner and due action; do not invent missing specialist skills or claim those stages are automated.
Example handoff
Assignment: Northbank backend engineer. Stage: screening. Completed: brief v2 and calibrated scorecard. Missing: incident-leadership evidence for Alex. Next action: recruiter reviews the screening report and chooses whether to invite Alex. Outreach: not sent. Booking: not created. No candidate stage change has been executed.
Completion
Report completed artifacts and confirmed actions, open decisions, failed or unavailable capabilities, and the next owner. Stop at the requested outcome; a full assignment remains open while agreed stages still require action.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.