tabular-review

Tabular review — one row per document, one column per data point, every cell cited to source. Built for M&A diligence ("review these 200 target contracts for change-of-control, assignment, and MAC clauses") but works for any batch review that needs a spreadsheet out the other end. Use when user says "tabular review", "review grid", "build a grid", "extract these fields from these contracts", "review these documents for X, Y, Z", "give me a spreadsheet of", "batch review", or points at a folder of documents and asks to compare them.

Quality

92%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Low

Low-risk findings worth noting

Quality

Content

85%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 well-structured, highly actionable instruction skill with a clear sequenced workflow, strong validation feedback loops for a batch operation, and clean one-level-deep progressive disclosure into real bundle files. The only weakness is repetition of the verbatim rule and the lead-not-finding reminder across multiple sections, which inflates token cost without adding guidance.

Suggestions

State the verbatim rule once authoritatively in Step 3 and cross-reference it from Step 4 and the "What this skill does not do" section instead of restating it in full each time.

Keep the "every cell is a lead, not a finding" reminder in the Step 6 Summary only and reference it from Purpose and Output safeguards to remove the triple repetition.

Tighten the Matter context paragraph, which is dense with plugin-internal path machinery that could be condensed or deferred to the practice-level CLAUDE.md it points to.

Dimension	Reasoning	Score
Conciseness	The body is mostly efficient and free of generic concept explanations, but the verbatim rule is restated fully in Step 3 and again in Step 4 and the "What this skill does not do" section, and the "every cell is a lead, not a finding" reminder recurs in Purpose, Step 6, and Output safeguards — tightening this repetition would move it to 3.	2 / 3
Actionability	Provides concrete, specific guidance throughout: a full YAML schema example, the typed column-system table, the exact row structure {value, state, quote, location}, specific notes keys (quote_unavailable, quote_mismatch), and concrete sample sizes (3–5 docs, 10% spot-check).	3 / 3
Workflow Clarity	Steps 0–6 are clearly sequenced with explicit validation checkpoints for this batch operation: a sample run before fan-out, schema confirmation, and a normalize pass with a re-read-and-compare spot-check feedback loop that downgrades and widens on mismatch.	3 / 3
Progressive Disclosure	The body is an overview that delegates detailed mechanics to real one-level-deep references (references/excel-output.md, references/gsheets-output.md, references/ma-diligence-columns.md, all present in the bundle), keeping the workflow and conceptual core inline while splitting out the output specs and column template.	3 / 3
	Total	11 / 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 specific, trigger-rich, and clearly answers both what the skill does and when to use it, with concrete domain framing that distinguishes it from adjacent skills. Voice is third-person/imperative ("Use when...") consistent with the rubric's good examples, so no voice penalty applies.

Dimension	Reasoning	Score
Specificity	Names multiple concrete actions — "one row per document, one column per data point, every cell cited to source" and reviewing contracts "for change-of-control, assignment, and MAC clauses" — rather than vague language.	3 / 3
Completeness	Explicitly answers both what ("Tabular review — one row per document, one column per data point, every cell cited to source") and when ("Use when user says...") with explicit triggers, matching the score-3 anchor.	3 / 3
Trigger Term Quality	Lists many natural phrases a user would actually say ("tabular review", "review grid", "build a grid", "give me a spreadsheet of", "batch review", "points at a folder of documents and asks to compare them"), with strong coverage of common variations.	3 / 3
Distinctiveness Conflict Risk	Occupies a clear niche (batch document review producing a cited spreadsheet) with distinct triggers; the M&A-diligence framing and specific trigger phrases make it unlikely to fire for the wrong skill.	3 / 3
	Total	12 / 12 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: anthropics/claude-for-legal
Path: corporate-legal/skills/tabular-review/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.