cold-start-interview

Run the cold-start interview — learns your privacy practice and writes CLAUDE.md from your policy, DPA template, and a reference PIA. Use on first run, when CLAUDE.md is missing or has placeholders, or when the user says "set up the privacy plugin", "onboard me", "configure privacy", or wants to re-run the interview or re-check integrations.

Quality

88%

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

Quality

Content

77%Weight 40%Scale 1-3

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

The body is highly actionable and well-sequenced with robust validation checkpoints, but it is verbose and keeps a large embedded template inline rather than using progressive disclosure. Tightening the conversational passages and extracting the profile template would lift both weak dimensions.

Suggestions

Move the embedded practice-profile markdown template (lines ~295-436) into a references/ file and reference it from the body, cutting the inline bulk and improving progressive disclosure.

Trim the conversational preamble paragraphs (e.g. "Why this matters", the attorney-referral block, and the repeated "you can change anything later" notes) to lean instructional prose that assumes Claude's competence.

Consolidate the duplicated re-run/change-later guidance (it appears in the Quick start path, the After writing section, and the closing notes) into a single stated-once block.

Dimension	Reasoning	Score
Conciseness	The ~490-line body is mostly efficient domain orchestration but padded with conversational monologues (e.g. the "Why this matters" paragraph, the attorney-referral passage, and repeated "you can change anything later" reassurances) that could be tightened without losing substance.	2 / 3
Actionability	Guidance is concrete and copy-ready: exact file paths, literal marker strings ("[PLACEHOLDER]", "<!-- SETUP PAUSED AT: -->"), specific interview questions, and a complete output profile template with exact slash-command invocations.	3 / 3
Workflow Clarity	Steps are clearly sequenced (1-6 overview, Parts 0-5) with explicit decision branches in the cold-start check and strong validation/feedback loops: probe integrations by actual MCP call success, verify legal facts before writing, and pause/resume with distinct [PENDING] vs [PLACEHOLDER] markers.	3 / 3
Progressive Disclosure	The skill is monolithic: a single SKILL.md embeds the full ~140-line practice-profile markdown template inline rather than splitting it into a referenced file, and there are no references/scripts/assets bundle files to offload detail.	2 / 3
	Total	10 / 12 Passed

Description

100%Weight 40%Scale 1-3

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 is strong: it states concrete actions, includes explicit natural-language triggers, and clearly answers both what the skill does and when to use it. It carves a distinct privacy-onboarding niche that is unlikely to collide with other skills.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions — "learns your privacy practice and writes CLAUDE.md from your policy, DPA template, and a reference PIA" — matching the anchor for several specific concrete actions rather than a single vague domain description.	3 / 3
Completeness	Explicitly answers both what ("learns your privacy practice and writes CLAUDE.md from your policy, DPA template, and a reference PIA") and when ("Use on first run, when CLAUDE.md is missing or has placeholders, or when the user says...") with an explicit trigger clause.	3 / 3
Trigger Term Quality	Includes natural user utterances such as "set up the privacy plugin", "onboard me", "configure privacy", and "re-check integrations" alongside state-based triggers, giving good coverage of phrases a user would actually say.	3 / 3
Distinctiveness Conflict Risk	The privacy-specific niche and the concrete CLAUDE.md-writing action, combined with triggers like "configure privacy" and "re-check integrations", make it unlikely to fire for unrelated skills.	3 / 3
	Total	12 / 12 Passed

Validation

81%

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 — 13 / 16 Passed

Validation for skill structure

Criteria	Description	Result
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning
relative_links	Relative link issues: 2 missing	Warning
referenced_paths_exist	Referenced path issues: 1 missing	Warning

	Total	13 / 16 Passed

Repository: anthropics/claude-for-legal
Path: privacy-legal/skills/cold-start-interview/SKILL.md
Commit: 4a6c651

Reviewed: 6 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.