setup-interview

Domain-by-domain interview to produce $CLAUDE_PROJECT_DIR/docs/project-context.json. Invoked once by the orchestrator; the orchestrator then conducts all turns directly using Read/Write/Edit — no agent dispatching.

Quality

63%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./.claude/skills/setup-interview/SKILL.md

Quality

Content

81%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The body is a well-sequenced, highly actionable instruction-only skill with strong validation checkpoints and feedback loops for its committed/persistent output, limited mainly by minor redundancy and inline reference-like material that could be split out.

Dimension	Reasoning	Score
Conciseness	The body is largely efficient and assumes Claude's familiarity with CRM, Supabase, and Vercel, with only minor redundancy (the cleanup elements appear in two tables and the output contract is stated in both the per-turn loop and a dedicated Output format section).	4 / 5
Actionability	Guidance is concrete and mostly copy-paste ready — literal Bash git commit command, exact Write paths, full JSON schema, and exact output tokens (VALIDATED / FAILED:) — with the minor gap that per-domain questions are given as topic bullets rather than verbatim user-facing question strings.	4 / 5
Workflow Clarity	The sequence (startup detection -> per-turn loop -> cleanup derivation -> consistency checks -> final validation -> commit -> VALIDATED) is explicit, with validation checkpoints, a feedback loop (failed check -> targeted INTERVIEW question -> fix), and a pre-emit Verification checklist, matching the anchor for clear sequence with explicit validation and checklists.	5 / 5
Progressive Disclosure	No bundle files exist, so the skill is self-contained with well-organized section headers; it is well-structured, though reference-like material (the ~50-line JSON schema and the cleanup plain-language table) is inlined rather than split into a reference file, leaving minor organization gaps.	4 / 5
	Total	17 / 20 Passed

Description

46%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

The description conveys a concrete what and a distinctive niche, but it is written from an internal orchestrator perspective rather than a user-trigger perspective, so it lacks natural trigger phrases and an explicit 'Use when...' clause.

Suggestions

Add an explicit 'Use when...' clause with natural user phrases such as 'Use when setting up a new project's CRM configuration or running the initial setup interview'.

Replace internal jargon ('orchestrator', 'agent dispatching') with user-facing trigger language so the skill reads as something a user would naturally request.

List 1-2 more concrete actions (e.g., 'derives cleanup of unused CRM features and commits a validated spec') to lift specificity from 1-2 actions toward several.

Dimension	Reasoning	Score
Specificity	Names a concrete domain ('Domain-by-domain interview') and one concrete action ('produce $CLAUDE_PROJECT_DIR/docs/project-context.json'), but does not list several specific actions, so it sits at the 1-2 concrete actions anchor rather than 4.	3 / 5
Completeness	It has a clear 'what' (domain-by-domain interview producing a JSON file) but the 'when' is only weakly implied via 'Invoked once by the orchestrator' rather than an explicit 'Use when...' user trigger, which caps completeness at 3 per the rubric guideline.	3 / 5
Trigger Term Quality	The keywords ('orchestrator', 'agent dispatching', 'Read/Write/Edit') are technical jargon a user would not naturally say; the only semi-natural term is 'interview', and the natural user phrases for triggering this skill are absent.	2 / 5
Distinctiveness Conflict Risk	It carves a clear niche (a one-time setup-interview producing project-context.json with no agent dispatching), making overlap with other skills unlikely, though the internal/technical framing leaves minor distinguishability gaps versus a fully user-triggered description.	4 / 5
	Total	12 / 20 Passed

Validation

100%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository: marmelab/atomic-crm
Path: .claude/skills/setup-interview/SKILL.md
Commit: 64e2163

Reviewed: 4 days ago

Table of Contents

Discovery Implementation Validation

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.