core

The analyst's built-in capabilities: data-inspection tools and the always-available actions (visualize, ask_user, delegate).

Quality

68%

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 ./py-src/data_formulator/analyst/skills/core/SKILL.md

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.

The content is highly actionable with concrete parameters, code patterns, and a clear decision workflow with feedback loops. Its main weaknesses are conciseness and progressive disclosure: it is a long monolithic file whose large reference tables would benefit from being split into separately linked files.

Suggestions

Move the Chart Type Reference, Semantic Type Reference, and Statistical Analysis Guide into separate reference files (e.g., references/chart_types.md) and link to them from SKILL.md so the overview stays lean.

Tighten the prose sections (e.g., the 'Actions' intro and 'Choosing what to do') — trim restated rationale and keep only the actionable decision rules.

Once reference material is externalized, keep SKILL.md focused on the tools, actions, and the question-classification workflow as the overview layer.

Dimension	Reasoning	Score
Conciseness	The body is dense (~290 lines) with large inline reference tables and prose that restates rationale; the chart/semantic reference material earns its place as a lookup, but the whole is mostly efficient yet could be tightened and partly externalized.	2 / 3
Actionability	Provides concrete, executable guidance throughout — exact parameter names, concrete pandas/DuckDB loading patterns, allowed-library lists, exact chart_type values and encodings, and per-action argument specs that are copy-paste ready.	3 / 3
Workflow Clarity	The 'Choosing what to do' section maps question types to action counts and enforces an explicit feedback loop ('observe the result and decide your next move', 'add the next chart only if it answers a gap raised by the previous one'), giving a clear sequenced decision flow.	3 / 3
Progressive Disclosure	Well-organized into clear sections, but everything is inline in one monolithic file — the Chart Type Reference, Semantic Type Reference, and Statistical Analysis Guide are large reference blocks that could live in separate files; no bundle files are provided.	2 / 3
	Total	10 / 12 Passed

Description

60%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 about the concrete tools and actions it covers, but reads in internal/engineering voice rather than natural user language and omits an explicit use-when trigger. Adding a user-facing 'Use when...' clause with natural trigger terms would lift the weaker dimensions.

Suggestions

Add an explicit 'Use when...' clause with natural user-facing trigger terms (e.g., 'Use when the user asks to analyze, visualize, or explore workspace data') so the skill is discoverable from how users actually phrase requests.

Replace internal jargon like 'always-available actions (visualize, ask_user, delegate)' with language users would say (e.g., 'chart data, ask clarifying questions, or hand work off to a peer agent').

Keep the concrete tool/action list but frame it around user outcomes so the what and when are both explicit in the description field itself.

Dimension	Reasoning	Score
Specificity	Lists multiple specific concrete actions — 'data-inspection tools and the always-available actions (visualize, ask_user, delegate)' — naming exact capabilities rather than vague abstractions.	3 / 3
Completeness	It clearly states what the skill does (built-in inspection tools and actions) but lacks any 'Use when...' clause or equivalent trigger guidance, so the 'when' is only implied via the separate when_to_use field.	2 / 3
Trigger Term Quality	The terms present ('visualize', 'ask_user', 'delegate', 'data-inspection tools') are internal action/tool names, not phrases a user would naturally say; common user-facing variations are missing.	2 / 3
Distinctiveness Conflict Risk	Naming the three specific actions gives it a recognizable niche, but as the 'always-on baseline' it is broadly scoped ('built-in capabilities', 'data-inspection tools') and could overlap with other analysis skills.	2 / 3
	Total	9 / 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: microsoft/data-formulator
Path: py-src/data_formulator/analyst/skills/core/SKILL.md
Commit: 5d4f7b3

Reviewed: 8 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.