worker-classification

Classify a proposed worker engagement — employee, IC, temp, or vendor — by running the applicable jurisdiction tests and flagging misclassification gaps between the intended arrangement and what the facts actually support. Prospective use only. Use when someone says "we want to bring on a contractor", "is this a vendor or a temp", "how should we classify this person", or describes a proposed working arrangement.

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.

A highly actionable, well-sequenced legal workflow with strong validation gates and concrete templates. Its main weaknesses are verbosity from restating the scope-mismatch warning and a monolithic inline body that could offload templates and gate blocks to referenced files.

Suggestions

Consolidate the scope-mismatch messaging: state the gate logic once and reference the verbatim block/banner rather than re-explaining the same reasoning in three places (gate prose, out-of-scope block, and SCOPE MISMATCH banner).

Extract the verbatim gate output blocks and the Step 5 output markdown template into reference files (e.g., outputs.md, gates.md) and link to them one level deep to improve progressive disclosure and reduce SKILL.md length.

Move the full Step 1 intake question bank into a reference file, keeping only a brief pointer inline, so the core workflow reads leaner while preserving the detail for when it is needed.

Dimension	Reasoning	Score
Conciseness	Mostly efficient procedural guidance, but the scope-mismatch reasoning is restated redundantly across the gate prose, the verbatim out-of-scope block, and the SCOPE MISMATCH banner — tightening would remove padding; not 3 because some tokens do not earn their place, not 1 because it does not explain concepts Claude already knows.	2 / 3
Actionability	Provides a concrete intake question bank, a scored test-application table template, exact output markdown, specific escalation triggers, and copy-ready verbatim gate blocks; not 2, which only offers incomplete or pseudocode-level guidance.	3 / 3
Workflow Clarity	Steps 1–5 are clearly sequenced with explicit validation gates (prospective-only hard gate, no-silent-supplement, consequential-action gate) and feedback loops for error recovery; not 2, which has implicit or missing checkpoints. This is an analysis workflow rather than a destructive/batch operation, so the cap does not apply.	3 / 3
Progressive Disclosure	No bundle files exist (references/, scripts/, assets/ absent), so disclosure is judged on structure; the ~400-line body keeps the full output template, verbatim gate blocks, and intake bank inline rather than split into referenced files — some structure but content that should be separate is inline. Not 3 because it is well over 50 lines with material that could be one level deep, not 1 because sections are well-organized with no nested-reference indirection.	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.

A strong, specific description that states concrete capabilities, natural trigger phrasings, and an explicit Use-when clause in third person. It cleanly answers both what and when with low conflict risk.

Dimension	Reasoning	Score
Specificity	Names multiple concrete actions — 'Classify a proposed worker engagement — employee, IC, temp, or vendor — by running the applicable jurisdiction tests and flagging misclassification gaps' — matching the multi-action anchor; not the level below, which only names a domain and some actions.	3 / 3
Completeness	Explicitly answers both what (classify via jurisdiction tests and flag gaps) and when via an explicit 'Use when...' clause with triggers; not 2, where the 'when' is only implied. Voice is third person ('Classify', 'Use when someone says'), so no specificity penalty applies.	3 / 3
Trigger Term Quality	Includes natural phrasings users would say — 'we want to bring on a contractor', 'is this a vendor or a temp', 'how should we classify this person' — giving good coverage; not score 2, which expects missing common variations.	3 / 3
Distinctiveness Conflict Risk	Occupies a clear worker-classification niche with distinct contractor/vendor/temp triggers unlikely to fire for unrelated skills; not 2, which still risks overlap with similar skills.	3 / 3
	Total	12 / 12 Passed

Validation

93%

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 — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	15 / 16 Passed

Repository: anthropics/claude-for-legal
Path: employment-legal/skills/worker-classification/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.