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.

71

Quality

89%

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

88%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 is lean, fully executable, and sequenced with explicit pre-checks, done-when criteria, and error fallbacks for both the run and teach paths. The one material weakness is progressive disclosure: the Learn half depends on references/docs/, which is absent from the bundle, so the referenced detail cannot actually be reached.

Suggestions

Ship the references/docs/ pages that SKILL.md and curriculum.md point to (index.mdx, cli/index.mdx, agents/*.mdx, etc.), or trim the references and curriculum table to only files that exist in the bundle.

State in the Teaching section what to do when references/docs/ is unavailable beyond the generic missing-page fallback, so the Learn path degrades gracefully instead of dead-ending at curriculum.md.

Consider noting in SKILL.md that the bundle is incomplete if docs are intentionally external, so a reader can distinguish a packaging gap from a broken reference.

DimensionReasoningScore

Conciseness

The body is lean and imperative throughout ("Execute `ywai …` yourself", "List before mutate. Confirm the id or name exists.", "Never enable or disable `core`"), with no explanation of concepts Claude already knows and every line carrying instruction. It matches the anchor of every token earning its place.

5 / 5

Actionability

Concrete copy-paste commands cover every mutation ("ywai mcp enable <id>", "ywai profile use <name>", "ywai groups disable <name>"), the expected outputs are quoted in "Done when" ("MCP … enabled", "Active profile: …"), and the teaching procedure is a numbered, specific sequence. This matches the fully-executable top anchor.

5 / 5

Workflow Clarity

Both workflows are fully sequenced with explicit validation and error feedback: pre-check ("List before mutate. Confirm the id or name exists."), post-check ("The command printed success … or you reported the error verbatim"), recovery for teaching ("If a page is missing or the read fails, say so and teach only what you successfully read"), plus the restart instruction after config changes. This matches the anchor with explicit validation steps and feedback loops.

5 / 5

Progressive Disclosure

The overview-to-reference design is right (SKILL.md stays lean and clearly signals "Open [references/curriculum.md](references/curriculum.md)"), but scoring against the actual bundle, the primary referenced content is missing: neither the body's "Read only `references/docs/`" nor curriculum.md's ~15 paths under `references/docs/` exist in the bundle — only references/curriculum.md is present. The disclosure chain breaks for the skill's entire Learn half, which is a more severe defect than the "not clearly signaled" anchor 3 example and lands noticeably below the midpoint.

2 / 5

Total

17

/

20

Passed

Description

87%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 states what the skill does (run and learn ywai across MCP, model profiles, agent groups, and doc tours) and supplies concrete trigger phrases, with the unique tool name keeping conflict risk minimal. It sits just below top marks on specificity and trigger quality because capabilities are a compressed noun list and triggers are command-style rather than covering natural-language variations.

DimensionReasoningScore

Specificity

"Run and learn ywai: MCP, model profiles, agent groups, doc tours" lists several concrete capability areas, matching the anchor for several specific actions with minor coverage gaps. It stays below the comprehensive anchor because the areas are named as compressed nouns without saying what is done in each (e.g. enable/disable MCP, switch profiles).

4 / 5

Completeness

The description explicitly answers both: what ("Run and learn ywai: MCP, model profiles, agent groups, doc tours") and when ("Trigger: ywai mcp, ywai profile, /learn-ywai") with concrete trigger phrases, matching the top anchor. The "Trigger:" clause is equivalent explicit trigger guidance, so the missing-'Use when' cap does not apply.

5 / 5

Trigger Term Quality

"Trigger: ywai mcp, ywai profile, /learn-ywai" gives directly relevant trigger phrases a user of this tool would naturally type. It falls short of the top anchor because natural-language synonyms and variations (e.g. "turn on the MCP", "change the model", "teach me ywai") are missing.

4 / 5

Distinctiveness Conflict Risk

The unique tool name "ywai" appears in both the capability statement and every trigger phrase, giving the skill a clear niche with minimal conflict risk with other skills. No generic phrasing that would overlap with unrelated skills is present.

5 / 5

Total

18

/

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.