CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

start-learning

One-time onboarding for the AI Engineering from Scratch curriculum (542 lessons, 20 phases). Interviews the learner, runs the placement quiz, and writes LEARNING.md, a persistent study plan the learn skill drives. Trigger phrases: "start learning", "set up the course", "begin the curriculum", "onboard me", "create my learning plan"

Quality

93%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Low

Low-risk findings worth noting

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.

A well-engineered onboarding workflow: concrete, executable guidance with explicit validation and confirmation gates around destructive actions, and a clear 4-step sequence. The main costs are token weight from repeated handoff scaffolding and route bookkeeping inlined in the main file rather than factored into a reference.

Suggestions

Factor the resume-routing route table (the seven state-file-to-owner mappings and their resolution rules) into a small reference file so the main body keeps only the decision algorithm.

Collapse the three parallel 'Focused <route> handoff' sections into one parameterized handoff pattern with a short route table (skill name, source JSON, state file), trimming the repeated 'do not run placement / do not create LEARNING.md' scaffolding.

DimensionReasoningScore

Conciseness

The body assumes Claude's competence — no concept explanations, every section carries non-obvious operational rules — but the three near-identical focused-handoff paragraphs (MCP, Agent Skills, Reverse Engineering) and the dense resume-routing block repeat structure that could be tightened. Not 5 because of that repeated handoff scaffolding; not 3 since almost all content earns its place.

4 / 5

Actionability

Guidance is fully concrete for an instruction-only skill: exact interview questions with offered options, a complete LEARNING.md template in a markdown block, precise validation rules ('Validate the phase is 0-19 and resolve its canonical name'), and exact three-line closing handoff. Not the 4 anchor because there are no missing key details.

5 / 5

Workflow Clarity

A clear 4-step sequence with explicit checkpoints and error-recovery loops: route resolution before any onboarding action, phase validation with a fallback ('if it does not resolve, list the 20 phases and ask them to pick'), and a confirmation gate before the destructive archive rename ('only after an explicit confirmation… Never delete or overwrite their history silently'). Not the 4 anchor because validation is explicit rather than implicit.

5 / 5

Progressive Disclosure

The skill has no bundle files, and the single body is well-sectioned and navigable with clear headers and a one-level-deep reference to ROADMAP.md (with a URL fallback). However, the ~180-line body inlines the 7-route resume-routing table and three parallel handoff sections that could live in a short reference file, which the 5 anchor would require split out.

4 / 5

Total

18

/

20

Passed

Description

96%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 strong description: concrete third-person actions, an explicit trigger-phrase list, and a clearly bounded one-time onboarding purpose. The only weakness is slight overlap risk between the 'start learning' trigger and the sibling learn skill that consumes its output.

DimensionReasoningScore

Specificity

The description lists multiple concrete actions covering the whole workflow — 'Interviews the learner, runs the placement quiz, and writes LEARNING.md, a persistent study plan' — in third-person voice with no vague padding. It is not the 4 anchor because there are no minor gaps in coverage.

5 / 5

Completeness

Both what ('Interviews the learner, runs the placement quiz, and writes LEARNING.md') and when ('Trigger phrases: …') are explicitly stated with concrete triggers, matching the 5 anchor exactly. The 4 anchor would require the when to be less explicit.

5 / 5

Trigger Term Quality

Five natural trigger phrases are given ('start learning', 'set up the course', 'begin the curriculum', 'onboard me', 'create my learning plan') — phrases a user would plausibly say, with good synonym variation. Not the 4 anchor because no common natural phrasing for this intent is obviously missing.

5 / 5

Distinctiveness Conflict Risk

The one-time onboarding niche is distinct and anchored to a specific curriculum, but 'start learning' is a phrase that could also plausibly trigger the closely related 'learn' lesson-delivery skill. Not 5 because of that minor sibling-skill overlap; not 3 because the dedicated triggers ('onboard me', 'set up the course') are clearly onboarding-specific.

4 / 5

Total

19

/

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

frontmatter_unknown_keys

Unknown frontmatter key(s) found; consider removing or moving to metadata

Warning

Total

15

/

16

Passed

Repository

rohitg00/ai-engineering-from-scratch

Path

skills/start-learning/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.