discovery-interview
Deep interview process to transform vague ideas into detailed specs. Works for technical and non-technical users.
49
23%
Does it follow best practices?
Impact
96%
1.23xAverage score across 3 eval scenarios
Advisory
Suggest reviewing before use
Optimize this skill with Tessl
npx tessl skill review --optimize ./.claude/skills/discovery-interview/SKILL.mdQuality
Discovery
7%Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.
This description is too vague to effectively guide skill selection. It fails to specify concrete actions, lacks natural trigger terms users would use, and provides no explicit guidance on when Claude should select this skill. The core concept of an interactive interview for spec generation is potentially distinctive but needs much more detail.
Suggestions
Add a 'Use when...' clause with trigger terms like 'spec', 'requirements', 'PRD', 'product requirements', 'feature spec', 'define requirements', 'flesh out idea', 'brainstorm feature'.
List specific concrete actions such as 'Asks clarifying questions to gather requirements, identifies edge cases, produces structured specification documents with user stories, acceptance criteria, and technical constraints'.
Specify the output format and domain more clearly, e.g., 'Generates product requirement documents, technical specifications, or feature briefs through an iterative Q&A process'.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | The description uses vague language like 'deep interview process' and 'transform vague ideas into detailed specs' without listing concrete actions. It doesn't specify what kind of specs, what the interview involves, or what outputs are produced. | 1 / 3 |
Completeness | The 'what' is vaguely described as transforming ideas into specs, but there is no explicit 'when' clause or trigger guidance. There is no 'Use when...' or equivalent, and the what itself is too abstract to be useful. | 1 / 3 |
Trigger Term Quality | The description lacks natural keywords users would say. Terms like 'deep interview process' are not phrases users would naturally use. Missing trigger terms like 'spec', 'requirements', 'brainstorm', 'plan', 'design doc', 'PRD', 'feature request', etc. | 1 / 3 |
Distinctiveness Conflict Risk | The concept of an interview-based spec generation process is somewhat distinctive, but 'transform vague ideas into detailed specs' could overlap with planning, brainstorming, or general project management skills. The lack of specificity about the domain or output format increases conflict risk. | 2 / 3 |
Total | 5 / 12 Passed |
Implementation
39%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
The skill provides a comprehensive and well-structured discovery interview framework with clear phases, good conflict detection, and a solid spec template. However, it is extremely verbose—explaining concepts Claude already knows (interviewing techniques, technology tradeoffs, how to talk to non-technical users) and embedding everything in a single monolithic file. The content would benefit enormously from aggressive trimming and splitting into referenced sub-files.
Suggestions
Cut content by at least 50%: remove the knowledge gap detection table, user type handling section, AskUserQuestion best practices, and example interview flow—Claude already knows how to interview effectively. Focus only on the phase structure, category list, completeness checklist, and spec template.
Split into multiple files: move the spec template to a separate SPEC_TEMPLATE.md, the category deep-dive questions to CATEGORIES.md, and keep SKILL.md as a concise overview with phase descriptions and references.
Remove explanatory content like 'Think of an API like a waiter' analogies and technology comparison examples (WebSockets vs SSE)—these teach Claude things it already knows rather than providing project-specific guidance.
Make AskUserQuestion calls consistent with actual tool syntax if this is a real tool, or clarify that these are illustrative patterns rather than executable invocations.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Extremely verbose at ~350+ lines. Contains extensive explanations Claude already knows (how to ask good questions, what WebSockets vs SSE are, analogies for non-technical users). The example interview flow, knowledge gap detection table, and user type handling sections explain interviewing fundamentals that Claude inherently understands. Much of this could be condensed to 1/3 the length. | 1 / 3 |
Actionability | Provides a structured interview process with concrete question examples, category breakdowns, and a spec template. However, the 'code' examples are pseudocode-like AskUserQuestion calls that aren't real tool invocations with proper syntax, and the guidance is more of a framework than executable steps. The spec template is concrete and useful. | 2 / 3 |
Workflow Clarity | The 7-phase process is clearly sequenced with explicit checkpoints: completeness checklist before spec generation, confirmation of understanding before finalizing, and clear decision points throughout. The conflict resolution phase and research loops provide good feedback mechanisms. The iteration rules set clear minimum thresholds. | 3 / 3 |
Progressive Disclosure | This is a monolithic wall of text with everything inline. The categories (A through H), example flows, best practices, user type handling, and spec template are all in one massive file. The spec template, category deep-dive details, and example interview flows could easily be split into referenced files. No external references are used. | 1 / 3 |
Total | 7 / 12 Passed |
Validation
90%Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.
Validation — 10 / 11 Passed
Validation for skill structure
| Criteria | Description | Result |
|---|---|---|
frontmatter_unknown_keys | Unknown frontmatter key(s) found; consider removing or moving to metadata | Warning |
Total | 10 / 11 Passed | |
- Repository
- parcadei/Continuous-Claude-v3
- Commit
d07ff4b
- Reviewed
Table of Contents
Is this your skill?
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.