sandbox-file-discovery-and-validation

Use shell tools safely in sandboxed environments by discovering files from real workspace roots instead of assumed user folders, then explicitly verifying output artifacts after generation.

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 ./benchmarks/gdpval/skills/sandbox-file-discovery-and-validation/SKILL.md

Quality

Content

77%

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 with a clear, well-validated workflow, but it is longer and more repetitive than necessary for a single-file skill with no external references. Consolidating repeated patterns would materially improve it.

Suggestions

Collapse the three restatements of the discovery and validation commands (Recommended workflow, Shell patterns to reuse, and Example end-to-end pattern) into a single canonical reference to remove repetition.

Trim the ~218-line monolith by merging the near-duplicate 'Prefer this / Avoid this' and 'Core principles' lists into one concise rules section.

If the skill grows further, split the reusable shell-pattern catalog into a references/ file and keep SKILL.md as a lean overview with a clear pointer.

Dimension	Reasoning	Score
Conciseness	There is no concept-padding, but the ~218-line body repeats the same discovery/validation commands and the prefer/avoid lists across 'Recommended workflow', 'Shell patterns to reuse', and the 'Example end-to-end pattern', so it could be tightened considerably.	2 / 3
Actionability	The body is dense with fully executable, copy-paste-ready commands such as `find . -type f -iname "*.xlsx"`, `mkdir -p output`, and `test -f ... && test -s ...`, with concrete examples for each phase.	3 / 3
Workflow Clarity	The 7-step workflow and 6-step end-to-end example are clearly sequenced with explicit validation checkpoints (sections 5 and 6) and a 'do not trust the exit code alone' feedback emphasis, matching the score-3 anchor.	3 / 3
Progressive Disclosure	No bundle files exist and the skill is a single well-sectioned file, but at ~218 lines with repetition it exceeds the under-50-line simple-skill allowance, so content that could be condensed or split remains inline.	2 / 3
	Total	10 / 12 Passed

Description

50%

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 what the skill does but reads as jargon-heavy and lacks an explicit use-trigger, leaving it uniformly adequate rather than strong. Its trigger phrasing would rarely match what a user actually says.

Suggestions

Add an explicit 'Use when...' clause naming concrete triggers (e.g. when files are not in Desktop/Documents/Downloads, or when locating inputs in an unfamiliar sandbox).

Replace skill-internal jargon ('workspace roots', 'assumed user folders') with natural user-facing terms ('find files', 'locate inputs', 'my folders are missing').

Narrow or de-emphasize the generic 'verifying output artifacts' framing so the description centers on the distinct sandbox-discovery niche and is less likely to conflict with output-generation skills.

Dimension	Reasoning	Score
Specificity	It names the domain (sandboxed shell environments) and two concrete actions ("discovering files from real workspace roots", "explicitly verifying output artifacts"), but stops short of the multiple distinct actions seen in a score-3 description.	2 / 3
Completeness	The "what" is clear, but there is no "Use when..." clause; the "when" is only implied by "in sandboxed environments", which per the judging guidelines caps completeness at 2.	2 / 3
Trigger Term Quality	Relevant keywords exist (files, sandbox, shell) but are framed as skill-internal jargon ("workspace roots", "assumed user folders") rather than natural user phrasing like "find my files" or "Downloads folder missing".	2 / 3
Distinctiveness Conflict Risk	The sandbox file-discovery niche is somewhat distinct, but the appended "verifying output artifacts after generation" is generic and could overlap with many output-producing skills.	2 / 3
	Total	8 / 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: HKUDS/OpenSpace
Path: benchmarks/gdpval/skills/sandbox-file-discovery-and-validation/SKILL.md
Commit: c5a9c4b

Reviewed: 2 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.