agentsociety-use-dataset

Use when external datasets need to be searched, inspected, or downloaded for experiments or analysis.

Quality

71%

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

Fix and improve this skill with Tessl

tessl review fix ./extension/skills/agentsociety-use-dataset/v1.0.0/SKILL.md

Quality

Content

57%

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

A well-structured, competently organized skill body with good progressive disclosure and a clear command reference. It loses points on missing full example invocations, implicit workflow validation, and minor redundancy around the README guidance.

Suggestions

Add at least one complete executable example invocation showing the full 'python .agentsociety/bin/ags.py <command> ...' form with realistic arguments.

Insert an explicit validation checkpoint in the workflow, e.g. after download run 'files <id>' / 'readme <id>' to verify contents before inspecting, and 'info <id>' to check for version updates.

Replace the dot-graph with a one-line arrow sequence and de-duplicate the README-first guidance so it appears once in the workflow rather than across Overview and two Common Mistakes rows.

Dimension	Reasoning	Score
Conciseness	Mostly lean with no concept over-explanation, but the dot-graph renders a trivial linear 5-step flow that a one-line arrow list would convey, and the 'read README first' guidance is repeated in Overview and two Common Mistakes rows.	2 / 3
Actionability	The Quick Reference table gives real commands with concrete flags and names the runner path, but no complete copy-paste invocation (e.g. 'python .agentsociety/bin/ags.py search --tags survey --limit 5') is shown, leaving examples incomplete.	2 / 3
Workflow Clarity	The browse -> readme -> download -> inspect -> update sequence is clearly ordered, but validation checkpoints are only implicit (the 'inspect' step) and no explicit verify-download or version-check feedback loop appears in the workflow itself.	2 / 3
Progressive Disclosure	SKILL.md is a concise overview that clearly signals two one-level-deep references (listing-guide.md, metadata-format.md), both real files, with detail appropriately split out — matching the well-organized overview anchor.	3 / 3
	Total	9 / 12 Passed

Description

85%

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 concise, action-oriented description with an explicit trigger clause and a clear niche. Its main weakness is trigger-term coverage: it uses formal verbs rather than the natural phrasings users actually say.

Suggestions

Add natural user-facing trigger phrases to the description, e.g. 'Use when the user says download dataset, find data, browse datasets, or search datasets'.

Prepend a short declarative capability sentence (e.g. 'Searches, inspects, and downloads datasets from the AgentSociety platform.') before the 'Use when' clause to separate 'what' from 'when'.

Dimension	Reasoning	Score
Specificity	Lists three concrete actions — 'searched, inspected, or downloaded' — applied to 'external datasets', matching the anchor for multiple specific concrete actions rather than vague language.	3 / 3
Completeness	Explicitly states what it does (search/inspect/download datasets) and when to use it via an explicit 'Use when ... for experiments or analysis' trigger clause, satisfying both what and when.	3 / 3
Trigger Term Quality	Covers relevant natural terms (search, inspect, download, datasets, experiments, analysis) but omits common user phrasings like 'find data', 'browse datasets', or 'download dataset' that the body itself lists.	2 / 3
Distinctiveness Conflict Risk	The 'external datasets' scope combined with consumption-only actions (search/inspect/download) carves a clear niche distinct from sibling skills like create-dataset or experiment-config.	3 / 3
	Total	11 / 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: tsinghua-fib-lab/AgentSociety
Path: extension/skills/agentsociety-use-dataset/v1.0.0/SKILL.md
Commit: 1832bcb

Reviewed: about 6 hours 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.