llm-wiki

The foundational knowledge distillation pattern for building and maintaining an AI-powered Obsidian wiki. Based on Andrej Karpathy's LLM Wiki architecture. Use this skill whenever the user wants to understand the wiki pattern, set up a new knowledge base, or needs guidance on the three-layer architecture (raw sources → wiki → schema). Also use when discussing knowledge management strategy, wiki structure decisions, or how to organize distilled knowledge. This is the "theory" skill — other skills handle specific operations (ingesting, querying, linting).

Quality

77%

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 ./.skills/llm-wiki/SKILL.md

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.

An authoritative, highly actionable schema document weakened by carrying too much inline: nearly everything a maintainer needs is copy-paste ready, but the 685-line body inlines content that belongs in the (underused) references directory, leaves two of three reference files unlinked, and cites a script that is not in the bundle. Internal structure and sequencing are clear throughout.

Suggestions

Move the Page Template, Paper Deep-Dive Template, confidence formula/source-quality table, and Importance Tiering sections into a reference file (e.g. references/SCHEMA.md) and link them one level deep, keeping SKILL.md to the architecture, principles, and config-resolution core — this addresses both the conciseness and progressive-disclosure gaps.

Fix the dangling and unused bundle references: link `references/karpathy-pattern.md` (the pattern this skill is 'Based on Andrej Karpathy's LLM Wiki architecture'), reference `references/WRITING.md` from the Writing Profile Resolution section, and either ship `scripts/manifest.py` or remove the `scripts/manifest.py migrate` invocation from the Source key contract section.

Deduplicate: delete the repeated `CODE_UNDERSTANDING_CODEGRAPH_BIN` env-var entry and consolidate the provenance-marker rules stated in three places (Provenance Markers, Page Template, Core Principles #4) into the single canonical section.

Dimension	Reasoning	Score
Conciseness	The ~40KB body is dense project-specific schema knowledge, not padding about concepts Claude already knows, but there is clear tightening available: `CODE_UNDERSTANDING_CODEGRAPH_BIN` appears twice verbatim, the extracted/inferred/ambiguous rules are restated across the Provenance, Page Template, and Core Principles sections, and manifest-migration edge cases run long. Mostly efficient with some removable redundancy — anchor 3, not 4.	3 / 5
Actionability	Fully executable guidance throughout: complete shell functions (`obsidian_wiki_config_dir`, `find_config`), exact commands (`obsidian-wiki memory sync INGEST source="papers/attention.pdf" pages_created=3`, `scripts/manifest.py migrate <vault> --dry-run --from-root <old-vault-root>`), copy-paste page and paper templates, a concrete confidence formula with a scoring-bucket table, and a full env-var table. No meaningful gaps; exceeds anchor 4.	5 / 5
Workflow Clarity	The Config Resolution Protocol is a well-sequenced 0→3 process with explicit failure handling ("tell the user it doesn't exist and list the available vaults... then stop"), and the lifecycle state machine names illegal transitions and how `obsidian-wiki lint` reports them. Not 5 because there is no consolidated end-to-end workflow with validate→fix→retry checkpoints — operational sequencing defers entirely to companion skills; not 3 because the sequences present carry explicit checkpoints.	4 / 5
Progressive Disclosure	`references/MEMORY.md` is real, one level deep, and well signaled, but scored against the actual bundle: `references/WRITING.md` and `references/karpathy-pattern.md` are never referenced from the body (the description cites Karpathy yet never links his pattern file), `scripts/manifest.py` is invoked but absent, and ~450 lines of template/formula/tiering detail that belongs in reference files are inlined. Some structure with real gaps — anchor 3.	3 / 5
	Total	15 / 20 Passed

Description

83%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: explicit and specific trigger clauses, a clear statement of what the skill is, and an unusually good explicit boundary against sibling operational skills. Main residual risk is overlap with the wiki-setup skill on the 'set up a new knowledge base' trigger.

Dimension	Reasoning	Score
Specificity	"building and maintaining an AI-powered Obsidian wiki", "set up a new knowledge base", "guidance on the three-layer architecture", "knowledge management strategy, wiki structure decisions" lists several specific capabilities. Not 5 because the actions are conceptual (a theory skill) rather than a comprehensive set of concrete operations; not 3 because coverage goes beyond 1-2 actions.	4 / 5
Completeness	Explicitly answers both: what ("the foundational knowledge distillation pattern... three-layer architecture (raw sources → wiki → schema)") and when ("Use this skill whenever the user wants to understand the wiki pattern, set up a new knowledge base... Also use when discussing knowledge management strategy..."). The when clause is explicit with concrete triggers, matching the anchor-5 example's structure.	5 / 5
Trigger Term Quality	Natural phrases users would say — "wiki pattern", "knowledge base", "wiki structure decisions", "organize distilled knowledge", "Obsidian". Not 5 because common synonyms and format triggers a user might actually say ("second brain", vault, Zettelkasten) are missing; not 3 because coverage is broad, not partial.	4 / 5
Distinctiveness Conflict Risk	"This is the 'theory' skill — other skills handle specific operations (ingesting, querying, linting)" explicitly delineates it from siblings, but "set up a new knowledge base" overlaps the wiki-setup sibling skill. Not 5 due to that overlap with a closely related skill; not 3 because the boundary statement and Karpathy framing give it a clear niche.	4 / 5
	Total	17 / 20 Passed

Validation

81%

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 — 13 / 16 Passed

Validation for skill structure

Criteria	Description	Result
skill_md_line_count	SKILL.md is long (686 lines); consider splitting into references/ and linking	Warning
relative_links	Relative link issues: 3 missing, 4 suspicious	Warning
referenced_paths_exist	Referenced path issues: 3 missing	Warning

	Total	13 / 16 Passed

Repository: Ar9av/obsidian-wiki
Path: .skills/llm-wiki/SKILL.md
Commit: 972de66

Reviewed: 6 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.