CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

learn-ywai

Run and learn ywai: MCP, model profiles, agent groups, doc tours. Trigger: ywai mcp, ywai profile, /learn-ywai.

Quality

85%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Passed

No findings from the security scan

SKILL.md
Quality
Evals
Security

Quality

Content

92%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.

An exemplary operational skill body: fully executable commands, explicit mutation safety rules, concrete done-when criteria, and a disciplined one-page-per-turn teaching procedure, all in lean directive prose. The one real defect is bundle-level: the references/docs/ directory the entire teaching half depends on is missing from the skill bundle as shipped.

Suggestions

Ship the references/docs/ directory (index.mdx, cli/, agents/, guides/, workflows/, panel/, configuration/, etc.) referenced by both SKILL.md and curriculum.md, or trim the teaching section and curriculum to reference only files that exist in the bundle.

Add a one-line note in the Teaching section telling the reader to consult references/curriculum.md for the topic-shortcut table, so the shortcut mappings are discoverable without opening the file blind.

DimensionReasoningScore

Conciseness

The body is lean and purely directive ("Execute ywai … yourself. Do not paste the command for the user to copy.", "disable keeps the config") with zero explanation of concepts Claude already knows and no padding — every token earns its place, matching anchor 5.

5 / 5

Actionability

Copy-paste-ready command blocks for all three mutation areas (ywai mcp list/enable/disable/auth, ywai profile list/use, ywai groups enable/disable), a profile name/effect table, success-output criteria in "Done when", and example invocations (/learn-ywai agentes) — fully executable coverage of the common cases, matching anchor 5.

5 / 5

Workflow Clarity

The run workflow has explicit validation checkpoints ("List before mutate. Confirm the id or name exists" plus the "Done when" success-or-report-error-verbatim criterion) and the teaching workflow is a numbered 5-step sequence with an explicit error-recovery fallback ("If a page is missing or the read fails, say so and teach only what you successfully read") — matching anchor 5's clear sequence with validation and feedback loops.

5 / 5

Progressive Disclosure

Signaling and layering are well designed — SKILL.md stays an overview, commands are inlined appropriately, and teaching detail is deferred to a single clearly linked index (references/curriculum.md) one level deep. However, the primary referenced content tree references/docs/ (with its .mdx pages, which both the body and curriculum.md depend on) is absent from the actual bundle, so navigation to the teaching material fails — a bigger defect than anchor 4's minor organization gaps, though the clear signaling keeps it above anchor 3's buried references.

3 / 5

Total

18

/

20

Passed

Description

73%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.

A terse, distinct description with an explicit trigger clause and tool-specific keywords that make false-positive triggering very unlikely. Its main weakness is that the capability statement names topics rather than concrete actions, leaving the what-it-does underspecified relative to the when.

Suggestions

Replace the generic verbs in the capability list with concrete actions, e.g. "Enable/disable MCP servers, switch model profiles, install agent groups, and teach guided doc tours of ywai."

Broaden the trigger list with the natural phrases the other two job areas would produce, such as "ywai groups", "agent groups", and "learn ywai".

State the operator explicitly (e.g. "Runs ywai CLI commands for the user...") so the what-clause reads as a capability, not just a topic list.

DimensionReasoningScore

Specificity

Names the domain plus four capability areas ("MCP, model profiles, agent groups, doc tours") but the actions themselves stay generic ("Run and learn") — no concrete verbs like enable/disable/switch. This matches anchor 3 (domain plus limited concrete actions) rather than 2 (bare domain) or 4 (several specific actions).

3 / 5

Completeness

Both parts are present: a what ("Run and learn ywai: MCP, model profiles, agent groups, doc tours") and an explicit when ("Trigger: ywai mcp, ywai profile, /learn-ywai") with concrete tokens. Below anchor 5 because the what is compressed — it lists topics rather than concrete actions — while clearly above anchor 3, which requires a missing or only implied when.

4 / 5

Trigger Term Quality

"Trigger: ywai mcp, ywai profile, /learn-ywai" gives natural terms a user of this tool would actually say, matching anchor 4. Not 5 because common variations are missing ("groups", "agent groups", "agente", "learn ywai", "doc tour").

4 / 5

Distinctiveness Conflict Risk

"ywai" is a unique product name and every trigger ("ywai mcp", "ywai profile", "/learn-ywai") is tool-specific, giving a clear niche with minimal conflict risk against any generic skill.

5 / 5

Total

16

/

20

Passed

Validation

93%

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

CriteriaDescriptionResult

referenced_paths_exist

Referenced path issues: 3 missing

Warning

Total

15

/

16

Passed

Repository

YoizenSA/dev-ai-workflow

Path

ywai/skills/learn-ywai/SKILL.md

Reviewed

Table of Contents

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.