Automotive VE AI Skills Kit

Automotive VE AI Skills Kit

A non-confidential portfolio project that converts automotive value engineering workflows into reusable AI Skills, scripts, reports, and human-review guardrails.

v0.4.5 Release GitHub Project Case Study 30-Day Plan Supplement Demo 中文 README
GitHub Actions test status
2
Public projects
7
AI Skills
18
Unit tests
2
Releases

Combined portfolio evidence: automotive VE workflow kit plus ecommerce operations workflow kit, both published with releases and CI-backed tests.

What was built

The project is designed as a proof of work for an AI productivity BP / Skill product manager working with an automotive value engineering team.

Reusable Skills

Six SKILL.md workflows cover scenario mining, VAVE opportunity drafting, manual writing, adoption feedback analysis, evidence-claim auditing, and community Skill benchmark adaptation.

Workflow Scripts

Five Python CLI tools score Skill candidates, generate adoption reports, check BOM or quotation readiness, audit portfolio claims, and rank which community Skill patterns can be responsibly adapted into VE workflows.

Release Quality

v0.4.5 includes sample CSV inputs, generated Markdown outputs, 18 unit tests, and GitHub Actions validation on both main and tag builds.

Supplementary proof

A second public demo shows the same workflow pattern in ecommerce operations: convert product, order, inventory, support, and review work into standard fields, weekly risk reports, and human-reviewed action queues.

Ecommerce Ops AI Workflow Kit Simulated data demo, not a real merchant or customer-data project.
  • 1 Skill for ecommerce operations review.
  • Python CLI turns 3 CSV inputs into a Markdown risk report.
  • Green GitHub Actions test workflow and v0.1.4 release.
v0.1.4 Release View repository

Workflow evidence

The operating model starts with input readiness and ends with adoption measurement. AI prepares structured drafts; business owners keep final authority.

  1. 0

    Check input readiness

    Block rows missing supplier, annual volume, tax status, or evidence source.

  2. 1

    Mine the workflow

    Extract repeated tasks, inputs, outputs, rework causes, and sensitive boundaries.

  3. 2

    Score candidates

    Rank scenarios by frequency, time cost, standardization, risk, data, benefit, and user willingness.

  4. 3

    Productize the Skill

    Write triggers, inputs, steps, output templates, review gates, and success metrics.

  5. 4

    Measure adoption

    Track usage, preparation time, rework, field completeness, and feedback-driven fixes.

  6. 5

    Audit claims

    Link every public claim to evidence status, source strength, safe wording, and honest boundaries.

  7. 6

    Adapt community patterns

    Benchmark public Skill structures, rewrite domain logic for VE, and keep attribution boundaries visible.

Reviewer case study

A short case study explains how workflow notes become AI Skills, scripts, Markdown outputs, and human-review guardrails. It is written for reviewers who want the reasoning path, not only a list of files.

From workflow notes to reusable AI Skills
  • Defines source inputs and blocked fields before drafting.
  • Shows how automotive VE and ecommerce operations share a transferable process pattern.
  • States what the portfolio does not claim before review questions arise.
Read case study

Verification links

These links are intended for reviewers who want to verify the portfolio without relying on resume wording. They point to repositories, releases, test workflows, and generated evidence.

Main repository Skills, Python CLI, examples, tests, bilingual README. Main release Pinned v0.4.5 release for the automotive VE workflow kit. Main test workflow Public GitHub Actions test history for the main project. Case study Reviewer-friendly explanation of the workflow-to-Skill conversion method. 30-day landing plan Onboarding plan for interviews, pilot selection, Skill MVP delivery, and adoption review. Supplement repository Ecommerce operations workflow kit showing transferable process work. Supplement release Pinned v0.1.4 release for the ecommerce operations workflow kit. Supplement test workflow Public GitHub Actions test history for the ecommerce supplement.

30-day landing plan

The project maps directly to an internal AI BP onboarding plan: learn the workflow, select low-risk pilots, ship Skill MVPs, and measure adoption before scaling.

Read full plan
Week 1Role interviews and scenario map
Week 2Skill backlog and pilot selection
Week 3Skill MVP and user trial
Week 4Adoption report and iteration plan

Honest boundaries

  • This is a non-confidential portfolio project, not a Seres internal project.
  • It does not use real supplier quotes, vehicle BOM data, or proprietary programs.
  • Public evidence is limited to repositories, releases, tests, generated examples, and documentation.
  • AI output is positioned as structured draft preparation, not final cost, supplier, quality, or engineering approval.