deep-learning-book

Study companion and working knowledge base for the Deep Learning textbook by Goodfellow, Bengio & Courville (MIT Press, 2016), read free at deeplearningbook.org. Indexes all 20 chapters, carries a 2016-to-2026 delta layer naming what the book got right, what was superseded (transformers, AdamW, diffusion, double descent) and what still holds, and ships four deterministic tools: a prerequisite-aware reading-path planner, a training-failure diagnostic, a capacity-and-regularization planner, and a parameter/FLOP/activation-memory calculator. Use when studying or teaching this book, planning a route through it, deciding whether a chapter's advice is still current, or translating its math into a training decision. It points at the official chapters — it never reproduces them.

Quality

80%

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

Quality

Content

71%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-organized, intelligent body that avoids restating basics and offers executable tooling, but its progressive disclosure is undermined by chapter and supporting-file links that point to files not present in the bundle, and it lacks explicit validation feedback loops for its diagnostic tools.

Suggestions

Ship the chapter files referenced by the Chapter Index (chapters/chNN-*.md) or remove/relabel those links so navigation does not dead-end.

Add the missing Supporting Files (glossary.md, patterns.md, cheatsheet.md) or drop them from the listing to avoid broken references.

Add an explicit validate→fix→retry feedback loop for the diagnostic tools (e.g., what to do when training_diagnostics.py or capacity_planner.py reports a failure) to raise workflow clarity.

Dimension	Reasoning	Score
Conciseness	Efficient and assumes Claude's competence — the Core Frameworks are dense original synthesis with no padding over basics Claude already knows. Minor trims possible: the source-book/companion framing is restated across frontmatter and body, and a few sentences are rhetorical rather than instructional, so it does not quite reach the lean-every-token-earns-its-place anchor.	4 / 5
Actionability	Provides concrete executable tool commands with real flags and example invocations (reading_path_planner.py, training_diagnostics.py, etc.) alongside actionable mental models. Minor gaps: chapter/topic routing points to chapter files that are not present in the bundle, and several frameworks are conceptual rather than copy-paste executable.	4 / 5
Workflow Clarity	The "How to Use" section gives a clear routing sequence (no-arg → topic → chNN → "is this still true?" → "where do I start?") and tools advertise typed exit codes and --help. It does not reach 5 because there are no explicit validate→fix→retry feedback loops for the batch/diagnostic tools, and it stays above 3 because the routing table is genuinely coherent.	4 / 5
Progressive Disclosure	Good section structure with valid one-level-deep references to references/.md (which exist), but the Chapter Index links to chapters/chNN-.md and Supporting Files links to glossary.md, patterns.md, cheatsheet.md — none of which exist in the bundle, so navigation is undermined by broken references and content that should be in separate files is referenced-but-absent.	3 / 5
	Total	15 / 20 Passed

Description

90%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, specific description that clearly states both what the skill does and when to use it, with concrete tool names and trigger phrases. Third-person voice is maintained throughout and over-claiming is avoided.

Dimension	Reasoning	Score
Specificity	Lists multiple specific concrete actions with comprehensive coverage — "Indexes all 20 chapters", a delta layer "naming what the book got right, what was superseded", and "ships four deterministic tools" each named with its function (reading-path planner, training-failure diagnostic, capacity-and-regularization planner, parameter/FLOP/activation-memory calculator).	5 / 5
Completeness	Explicitly answers both what (companion, index of 20 chapters, delta layer, four tools) and when ("Use when studying or teaching this book...") with concrete trigger phrases, satisfying the highest anchor.	5 / 5
Trigger Term Quality	Has natural trigger phrases ("Use when studying or teaching this book, planning a route through it, deciding whether a chapter's advice is still current, or translating its math into a training decision") but relies on full-sentence triggers rather than keyword/synonym coverage, missing common variations like "Deep Learning textbook" or "Goodfellow".	4 / 5
Distinctiveness Conflict Risk	Occupies a clear niche as a companion to a specific named textbook with named tools and explicit routing away from sibling skills, giving it distinct triggers and minimal conflict risk.	5 / 5
	Total	19 / 20 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
relative_links	Relative link issues: 23 missing	Warning

	Total	15 / 16 Passed

Repository: alirezarezvani/claude-skills
Path: engineering/deep-learning-book/skills/deep-learning-book/SKILL.md
Commit: 19392f7

Reviewed: ~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.