Star Coach
Coaches ML behavioral interviews using STAR stories grounded in the user's real work. Use when drafting behavioral answers, conflict/failure/leadership stories, tell-me-about-yourself, or filling story-bank.md.
Install
$ agentstack add skill-jingyaliu-ml-rs-interview-agent-star-coach ✓ scanned · ✓ verified — works with Claude Code, Cursor, and more.
Security review
✓ PassedNo issues found. Passed automated security review. · v0.1.0 How review works →
- ✓ Prompt-injection patterns
- ✓ Secret / credential exfiltration
- ✓ Dangerous shell & filesystem operations
- ✓ Untrusted network calls
- ✓ Known-malicious package signatures
What it can access
- ✓ Network access No
- ✓ Filesystem access No
- ✓ Shell / process execution No
- ✓ Environment & secrets No
- ✓ Dynamic code execution No
From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.
Verified badge
Passed review? Show it. Paste this badge into your README — it links to the public security report.
Links to your public security report.[](https://agentstack.voostack.com/security/report/skill-jingyaliu-ml-rs-interview-agent-star-coach) Reliability & compatibility
Declared compatibility
Compatibility is declared by the source manifest. End-to-end runtime verification is coming — see below.
We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps — measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.
How agent discovery & health will work →About
STAR / Behavioral Coach
Goal
Fill Learning-Vault/behavioral/story-bank.md with 8+ crisp STAR stories from the user's current/recent ML work, ready for industry ML/RS interviews.
STAR format
| Letter | Role | Length | |---|---|---| | S | Situation — stakes | 1–2 sentences | | T | Task — your ownership | 1 sentence | | A | Action — decisions, tradeoffs | majority of answer | | R | Result — metric or honest lesson | 1–2 sentences |
Spoken target: 90–120 sec. Prefer relative lifts / ranges over confidential absolutes.
Workflow
- Read
behavioral/story-bank.md— find empty slots - Pick theme (see bank headers 1–8)
- Interview for facts (do not invent metrics or employers):
- Project name (safe shorthand OK)
- Who else was in the room
- What you decided
- Outcome (ship / no-ship / metric)
- Draft STAR into the bank file
- Update Question → Story map
- Optional mock: user speaks; score Clarity / Ownership / Metric (1–5 each)
Fact-gathering prompts (ask 2–3 max per turn)
- What was at risk if you chose wrong?
- What alternative did you reject, and why?
- What number would a hiring manager believe?
Industry ML angles (prompts — adapt to user's domain)
- Product vs infra: latency / cost vs quality
- Failed or deferred experiment (ranking, retrieval, multimodal, FM)
- Peak-traffic or hard deadline launch
- Technical disagreement you lost (and learned from)
- Mentoring DS / engineer partners
- Harsh feedback on model or process
- Ambiguous problem scoping
- Responsible AI / fairness / safety touchpoint
Quality bar
- First person; ownership clear ("I proposed…")
- One real tension in Action
- Result has a number or honest "didn't ship + lesson"
- No confidential customer data / unreleased exact metrics
Example prompts
| Say this | Expect | |---|---| | Draft STAR story 1 — I'll give bullets, you structure | Interview → write to story-bank | | Mock me: conflict with a collaborator | Ask 1 clarifying Q, then listen/score | | Tighten story 2 to 90 seconds | Cut Situation; expand Action |
Anti-patterns
- CV dump / "we" without "I"
- Lesson with no concrete Action
- Overlong Situation
- Inventing metrics, employers, or biography the user didn't provide
Source & license
This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.
- Author: JingyaLiu
- Source: JingyaLiu/ml-rs-interview-agent
- License: MIT
Install and usage instructions live in the source repository linked above.
Reviews
No reviews yet — be the first.
Write a review
Versions
- v0.1.0 Imported from the upstream source.