@zonease/aiworker-cli 0.13.2 → 0.14.0

package/official-apps/aiworker-hr/skills/candidate-profile/SKILL.md

@@ -0,0 +1,32 @@
+---
+name: candidate-profile
+description: Build and update a source-backed Candidate People Profile for hiring review.
+capabilities:
+  - hr-profile
+  - recruiting
+  - evidence-review
+---
+
+# Candidate Profile
+
+Use this skill when the workspace profile represents a candidate or candidate
+pool member.
+
+## Profile Contract
+
+- Treat `README.md` as the accepted Candidate Profile.
+- Treat files under `artifacts/` as proposed profile changes until review.
+- Keep confirmed facts, missing evidence, weak signals, and next HR actions
+  separate.
+- Do not infer protected-class attributes, personal judgments, or employment
+  commitments.
+
+## Output Shape
+
+For proposed updates, write a markdown artifact with:
+
+1. Current candidate summary
+2. Role-relevant evidence
+3. Missing or conflicting evidence
+4. Hiring risks and compliance notes
+5. Human reviewer next actions

package/official-apps/aiworker-hr/skills/evidence-screening/SKILL.md

@@ -0,0 +1,32 @@
+---
+name: evidence-screening
+description: Screen HR evidence for source quality, role relevance, missing facts, and risk.
+capabilities:
+  - hr-evidence
+  - candidate-screen
+  - risk-review
+---
+
+# Evidence Screening
+
+Use this skill to evaluate resumes, ATS packets, interview notes, employee
+records, or lifecycle touchpoints before they influence a People Profile.
+
+## Screening Standard
+
+- Tie every positive or negative signal to source evidence.
+- Mark weak, stale, missing, or conflicting evidence explicitly.
+- Separate role-related criteria from unsupported personal judgments.
+- Keep raw sensitive evidence out of durable profile history unless reviewed.
+- Prefer descriptor references over copying private evidence bodies.
+
+## Output Shape
+
+Create an evidence matrix with:
+
+1. Evidence source
+2. Role or lifecycle relevance
+3. Confidence
+4. Missing information
+5. Risk note
+6. Recommended human follow-up

package/official-apps/aiworker-hr/skills/hiring-risk-review/SKILL.md

@@ -0,0 +1,33 @@
+---
+name: hiring-risk-review
+description: Review HR profile proposals for compliance, privacy, bias, and unsupported claims.
+capabilities:
+  - hiring-risk
+  - compliance-review
+  - privacy-review
+---
+
+# Hiring Risk Review
+
+Use this skill before a proposed HR artifact is promoted into the accepted
+People Profile.
+
+## Risk Review Rules
+
+- Flag protected-class inference, proxy signals, and unsupported personal
+  judgments.
+- Flag copied sensitive evidence that should stay behind a descriptor or source
+  reference.
+- Flag employment commitments, compensation claims, or hiring decisions that are
+  not explicitly human-approved.
+- Separate blocking risks from advisory improvements.
+
+## Output Shape
+
+Return:
+
+1. Verdict recommendation: pass, warn, fail, or needs_review
+2. Blocking findings
+3. Advisory findings
+4. Privacy and evidence retention notes
+5. Required human decision before profile promotion

package/official-apps/aiworker-hr/skills/interview-brief/SKILL.md

@@ -0,0 +1,31 @@
+---
+name: interview-brief
+description: Draft evidence-backed interview briefs, focus areas, and scorecard guidance.
+capabilities:
+  - interview-planning
+  - recruiting
+  - scorecard
+---
+
+# Interview Brief
+
+Use this skill when preparing an interview plan from a Candidate Profile or
+candidate evidence packet.
+
+## Interview Guidance
+
+- Use role-relevant evidence and open questions from the profile.
+- Turn weak or missing evidence into interview focus areas.
+- Avoid protected-class inference and unsupported personal conclusions.
+- Keep the final hiring decision explicitly human-owned.
+
+## Output Shape
+
+Draft:
+
+1. Interview objective
+2. Evidence-backed focus areas
+3. Questions mapped to role criteria
+4. Scorecard guidance
+5. Risks or missing evidence to verify
+6. Reviewer notes for the human panel

package/official-apps/aiworker-hr/skills/profile-update-proposal/SKILL.md

@@ -0,0 +1,29 @@
+---
+name: profile-update-proposal
+description: Convert HR session evidence into a reviewable profile update proposal.
+capabilities:
+  - hr-profile
+  - artifact-proposal
+  - review-prep
+---
+
+# Profile Update Proposal
+
+Use this skill when a session should produce a proposed change for the accepted
+People Profile.
+
+## Rules
+
+- Read `README.md` first to understand the accepted profile baseline.
+- Write proposed changes under `artifacts/<sessionId>/`.
+- Do not update `README.md` directly unless the operator explicitly asks for a
+  reviewed profile revision workflow.
+- Preserve source references, open questions, and human decision ownership.
+
+## Proposal Checklist
+
+- What changed since the accepted profile?
+- Which claims are backed by explicit evidence?
+- Which facts are still missing?
+- Which risks need HR or legal review?
+- What exact review decision is needed before promotion?

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@zonease/aiworker-cli",
-  "version": "0.13.2",
+  "version": "0.14.0",
   "description": "AIWorker CLI — local Host and vertical Soul workspace runtime",
   "license": "MIT",
   "type": "module",