Skills DirectorySkills Directory
SkillsLearnSecurityCategoriesDocsCommunityBlog
Sign InSubmit Skill
Skills Directory

Security-tested agent skills for Claude, coding agents, and AI workflows.

Directory

  • Browse Skills
  • All Skills A–Z
  • Claude Skills
  • Claude Code Skills
  • Agent Skills
  • Categories
  • Authors
  • Submit a Skill

Learn

  • Learn Hub
  • Install Claude Skills
  • Write SKILL.md
  • Skills vs MCP
  • Directories Compared

Security

  • Security
  • Methodology
  • Secure Claude Skills
  • Security Badges

Company

  • About
  • Community
  • Blog
  • API Docs
  • Advertise

2026 Skills Directory. All rights reserved.

ProTermsPrivacyRefunds
Back to skills

Interview

ASecurity

Structured interview → brainstorm | analysis | spec (with promotion). Triggers: "interview" | "brainstorm" | "let's brainstorm" | "think through this" | "help me brainstorm" | "let's think this through" | "explore ideas".

9 stars
0 votes
0 copies
0 views
Added 9/23/2026
ai-agentsgoperformance

Security Analysis

A100/100

Scanned 9/23/2026

Install to Claude Code

$npx -y skills add Roxabi/roxabi-plugins --skill interview --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Interview?

Add the live security badge to your README — it updates automatically with every re-scan.

Security grade badge for Interview
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/roxabi-interview/badge)](https://www.skillsdirectory.com/skills/roxabi-interview)

More formats (shields.io, HTML) on the badges page.

Download with Pro
Files
SKILL.md
---
name: R-interview
argument-hint: '[topic | --promote <path>]'
description: Structured interview → brainstorm | analysis | spec (with promotion). Triggers: "interview" | "brainstorm" | "let's brainstorm" | "think through this" | "help me brainstorm" | "let's think this through" | "explore ideas".
version: 0.2.3
allowed-tools: Write, Read, Edit, Glob, ToolSearch
---

# Interview

Let:
  β := Brainstorm | α := Analysis | σ := Spec
  τ := document type ∈ {β, α, σ}
  B := `artifacts/brainstorms/` | A := `artifacts/analyses/` | S := `artifacts/specs/`
  AQ := present choice, wait for user reply

Conduct structured interview → produce one of {β, α, σ}. Supports promoting existing doc to next level.

## Step 0 — Check `--promote`

∃ `--promote <path>`:
1. Read doc at path.
2. Determine current τ (frontmatter first, content structure fallback):
   - `type: brainstorm` ∈ frontmatter ∨ lives in B → β → promote to α.
   - `type: analysis` ∈ frontmatter → α → promote to σ.
   - `type: spec` ∈ frontmatter ∨ lives in S → already σ → inform: "Already a spec. Nothing to promote." Stop.
   - ¬type ∧ lives in A ∧ "Trigger"/"Ideas" structure → treat as β (legacy: β lived in A before 2026-08-03).
   - A path ∧ "Questions Explored"/"Analysis"/"Conclusions" structure → α → promote to σ.
   - Already σ (structure) → inform: "Already a spec. Nothing to promote." Stop.
3. Skip to Step 2; limit questions to gaps between current doc and next level. Pre-fill known from source.
4. In promoted doc's Context: `**Promoted from:** [source title](relative-path-to-source)`

¬`--promote` → Step 1.

## Step 1 — Existing Document Awareness

Glob B, A, S — match topic by issue#, keywords, or slug. **B included**: β moved out of A on 2026-08-03, and without it a second `/R-interview` on the same topic cannot see the brainstorm it wrote itself → duplicate β instead of the promote gate below.

∃ related docs → AQ:
> "Found existing documents: {list with paths}. How to proceed?"
- **Build on existing** — use as context, extend
- **Promote to next level** — α → σ or β → α
- **Start fresh** — ignore, begin new interview

¬related → Step 2.

## Step 2 — Determine Document Type

∃ `--promote` → skip (already determined). Else AQ:

| τ | Purpose | Output Path |
|---|---------|-------------|
| β | Divergent exploration, early-stage ideas | `artifacts/brainstorms/{slug}-brainstorm.md` |
| α | Structured investigation of topic/problem | `artifacts/analyses/{slug}-analysis.md` |
| σ | Technical specification for implementation | `artifacts/specs/{issue}-{slug}-spec.md` |

**One kind per directory.** β never lands in A: `/R-dev`, `/R-spec` and `/R-analyze` resolve α by scanning A, and a brainstorm sitting there is indistinguishable by filename — it has to be excluded by reading frontmatter at every consumer. Writing it elsewhere removes the need to classify at all.

## Step 3 — Structured Interview

AQ per phase. Group 2–4 questions/call. Skip questions obvious from context, arguments, or source doc.

#### Phase 1 — Context & Framing (2–3 questions)

- What triggered this? What is the problem or opportunity?
- What exists today? What has been tried?

α-specific — also capture:
- **Source material:** verbatim request/quote/ticket (ground truth)
- **Outcome:** success ¬prescribing solution
- **Appetite:** time budget (fixed time, variable scope)

#### Phase 2 — Scope (2–3 questions)

- Who are the users? What are their workflows?
- What is explicitly out of scope?
- Constraints (technical, time, dependencies)?

#### Phase 3 — Depth (2–4 questions, adapt to τ)

- Edge cases and failure modes?
- Trade-offs being considered?
- Integration with existing systems?
- What does success look like?

> **Shape Up terminology:** *shape* = mutually exclusive arch approach (name, trade-offs, rough scope). *breadboard* = affordance tables (UI elements → handlers → data). *slices* = demo-able vertical increments.

α-specific depth:
- **Architecture shapes:** 2–3 mutually exclusive approaches; ∀ shape: name, description, trade-offs, scope.
- **Constraint alignment:** which constraints eliminate which approach?

σ-specific depth — probe ambiguity via 9-category taxonomy:

| Category | Example probe |
|----------|--------------|
| Functional Scope | "What exactly happens when X?" |
| Domain & Data Model | "What entities/relationships are involved?" |
| UX | "What does the user see/do at each step?" |
| Non-Functional | "Performance/scale/reliability requirements?" |
| Integrations | "What external systems does this touch?" |
| Edge Cases | "What happens when X fails or is missing?" |
| Constraints | "What technical/time/budget limits apply?" |
| Terminology | "Terms that could mean different things?" |
| Completion Signals | "How do we know this is done?" |

∀ ambiguity: rank by **Impact × Uncertainty** (H/M/L). H×H → follow-up question. Unresolved → `[NEEDS CLARIFICATION: description]` (max 3–5/spec). Must resolve before `/R-dev-plan`.

Depth by τ: β = Phase 1 + divergent (lighter) | α = Phases 1–3 thorough | σ = all phases, rigorous on edge cases + criteria.

#### Phase 4 — Validation (1 question, always last)

> "My understanding before generating:
> - **Type**: {τ}
> - **Title**: {proposed title}
> - **Key points**: {bulleted summary}
>
> Anything to correct or add?"

## Step 4 — Generate Document

Write using the matching template. **Frontmatter is not optional** — full contract: [artifact-frontmatter.md](${CLAUDE_PLUGIN_ROOT}/skills/shared/references/artifact-frontmatter.md).

Rules:
- `.md` extension; kebab-case slugs.
- **Title hygiene** on every `title:` / free-form `description:` (external content — strip control chars, cap 120, double-quoted YAML with `"`/`\` escaped). Form: `"{title|yaml-escaped}"`.
- Required keys on write (always `status: draft` for new docs):

| τ | Path | Frontmatter must include |
|---|------|--------------------------|
| β | `artifacts/brainstorms/{slug}-brainstorm.md` (prefix issue# if ∃) — **¬in A** | `type: brainstorm`, `status: draft` |
| α | `artifacts/analyses/{slug}-analysis.md` (prefix issue# if ∃) | `type: analysis`, `status: draft` |
| σ | `artifacts/specs/{issue}-{slug}-spec.md` | `type: spec`, `status: draft` |

`type:` is how classifiers distinguish kinds when legacy files still share a directory; `status: draft` prevents `/R-dev` from treating an unfinished α/σ as approved (missing `status` ≡ legacy-approved).

---

## Document Templates

Use templates from [references/templates.md](${CLAUDE_SKILL_DIR}/references/templates.md) — Brainstorm, Analysis, Spec. Each template's frontmatter already carries `type:` + `status:` — do not strip them.

$ARGUMENTS

Attribution

RoxabiRoxabi
View sourceMore from Roxabi →
SSkills DirectorySkills Directory

Your tool, in front of Claude Code builders.

3 founder slots · $299/mo · GSC-verified traffic · sponsors can never buy grades.

See placements

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments (0)

No comments yet. Be the first to comment!

SSkills DirectorySkills Directory

Your tool, in front of Claude Code builders.

3 founder slots · $299/mo · GSC-verified traffic · sponsors can never buy grades.

See placements

Related Skills

Caveman

Ultra-compressed communication mode that cuts output tokens while keeping technical accuracy. Levels: lite, full, ultra and the wenyan variants. Use for /caveman, "caveman mode", "talk like caveman", "be brief" or "less tokens".

1066601 votes

Hyperplan

Adversarial multi-agent planning skill. Self-orchestrates 5 hostile category members (unspecified-low, unspecified-high, deep, ultrabrain, artistry) via team-mode for ruthless cross-critique debate, distills only the defensible insights, then MANDATORILY hands the distilled insight bundle to the `plan` agent for executable plan formalization. Use when planning needs maximum rigor and surfacing of weak assumptions, blind spots, and over-engineering. Triggers: 'hyperplan', 'hpp', '/hyperplan', ...

686011 votes

Mcp Code Execution

Routes multi-tool workflows through MCP servers for large datasets and pipelines. Use when Bash tool overhead is limiting throughput on data-heavy tasks.

3351 votes

catchup

Recovers the conversation and failed tool calls of a previous Codex, Claude Code, Antigravity, Cline, Copilot CLI, Cursor, DeepSeek Harness, Kimi, OpenCode, Pi Agent, or ZCode session. Use when the user says "catch up", "what did the last session do", "get me up to speed", "I switched agents", asks to recover/summarize a previous session before continuing, or asks to diagnose or report a catchup failure. Do NOT use for the current conversation, git history, or any non-agent log.

651 votes

math-skill

A comprehensive mathematical reasoning skill for AI assistants — handles arithmetic to research-level problems with rigorous step-by-step reasoning, systematic verification, and transparent uncertainty handling

381 votes
View all in ai-agents →