reliable-excel-workbook-creation

Use run_shell with cd and heredoc syntax for complex Excel operations when read_file fails on XLSX or execute_code_sandbox fails on multi-step workbook creation

Quality

65%

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 ./benchmarks/gdpval/skills/reliable-excel-workbook-creation/SKILL.md

Quality

Content

68%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 delivers fully executable, copy-paste-ready Excel examples with clear section structure and no unnecessary concept explanations. Its weaknesses are structural redundancy across three restated sections and a lack of validation/verification steps for destructive file operations like overwriting an existing workbook.

Suggestions

Collapse the Step-by-Step Instructions, Why This Works table, and Best Practices sections into a single concise section to remove restated heredoc/absolute-path points.

Add a validation checkpoint to the read/modify example (e.g., back up the existing file before overwriting, or reload the workbook after save to confirm it opens), since overwriting is a destructive operation.

Show explicit error handling (try/except around save) rather than only mentioning it as a best practice, to give a concrete feedback loop for batch/destructive operations.

Dimension	Reasoning	Score
Conciseness	The code blocks are lean and do not over-explain concepts Claude already knows, but the Step-by-Step Instructions, Why This Works table, and Best Practices sections largely restate the same heredoc/absolute-path points, adding padded redundancy that could be tightened.	3 / 5
Actionability	All examples are fully executable, copy-paste ready Python scripts with imports, workbook creation, data population, formatting, and save, covering the common cases (multi-sheet creation, read/modify, formulas).	5 / 5
Workflow Clarity	A sequence is present (cd, heredoc, write script, save with absolute path), but validation checkpoints are missing — the 'Read and Modify Existing Workbook' example destructively overwrites a file with no backup/verification, and no example confirms the saved file is valid, capping this dimension at 3 per the batch/destructive-operations rule.	3 / 5
Progressive Disclosure	The single-file skill is well-organized with clear section headers and easy navigation, with no nested references; it loses a point because the redundant sections slightly inflate length for a single-pattern skill that could be tighter.	4 / 5
	Total	15 / 20 Passed

Description

62%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 explicitly covers both what to do and when to use it with concrete, domain-specific keywords, making it reasonably distinct and actionable. Its main weaknesses are trigger framing oriented around internal tool failures rather than natural user requests, and a lack of enumerated Excel capabilities.

Suggestions

Lead with a declarative capability statement (e.g., 'Creates and modifies Excel .xlsx workbooks via openpyxl') before the fallback trigger, mirroring the good examples' structure.

Reframe the 'when' clause around user-facing triggers (e.g., 'Use when creating multi-sheet workbooks, adding formulas, or formatting Excel files') in addition to the tool-failure conditions.

Add common synonyms such as 'spreadsheet' and '.xlsx' to broaden natural trigger-term coverage.

Dimension	Reasoning	Score
Specificity	It names the domain ('complex Excel operations', 'multi-step workbook creation') and a concrete technique ('run_shell with cd and heredoc syntax'), but does not enumerate the actual Excel capabilities (create, populate, format, save), so coverage is not comprehensive.	3 / 5
Completeness	Both 'what' (use run_shell with cd and heredoc for Excel operations) and 'when' (when read_file fails on XLSX or execute_code_sandbox fails on multi-step workbook creation) are explicitly stated with concrete triggers, though the 'when' is tool-failure-oriented rather than user-intent-oriented.	4 / 5
Trigger Term Quality	Relevant natural keywords are present ('Excel', 'XLSX', 'workbook'), but common synonyms like 'spreadsheet' are missing and the triggers are framed around tool failures ('read_file fails', 'execute_code_sandbox fails') rather than phrases users naturally say.	3 / 5
Distinctiveness Conflict Risk	The fallback-tooling niche (run_shell/heredoc when other tools fail on XLSX) is fairly distinct with specific triggers, with only minor overlap risk against a general Excel skill.	4 / 5
	Total	14 / 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: HKUDS/OpenSpace
Path: benchmarks/gdpval/skills/reliable-excel-workbook-creation/SKILL.md
Commit: 3827781

Reviewed: 24 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.