add-karpathy-llm-wiki

Add a persistent wiki knowledge base to a NanoClaw group. Based on Karpathy's LLM Wiki pattern. Triggers on "add wiki", "wiki", "knowledge base", "llm wiki", "karpathy wiki".

Quality

76%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Low

Low-risk findings worth noting

Fix and improve this skill with Tessl

tessl review fix ./.claude/skills/add-karpathy-llm-wiki/SKILL.md

Quality

Content

75%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-structured, actionable setup workflow with strong idempotency and overwrite-safety guidance. It is held back from a top score by minor verbosity in explanatory notes and the lack of an explicit post-setup verification loop.

Suggestions

Tighten the URL-handling note by dropping the explanation of WebFetch's behavior and stating only the directive (use bash/curl or agent-browser for full-text retrieval).

Add an explicit Step 6.5 verification checkpoint (e.g., confirm the wiki group responds to a test source and that wiki/index.md and wiki/log.md exist) to close the validate-fix-retry loop.

Move the lengthy ingest-discipline bullet and URL-handling note into a referenced file (e.g., references/ingest-discipline.md) to improve progressive disclosure.

Dimension	Reasoning	Score
Conciseness	The body is largely lean and action-oriented with concrete commands, but the URL-handling note explains WebFetch behavior Claude already knows and includes light editorializing ("this is critical — it's what turns the agent into a wiki maintainer").	4 / 5
Actionability	It supplies executable commands (mkdir -p, curl -sLo, launchctl kickstart, systemctl --user restart) and concrete AskUserQuestion options, with only minor gaps where placeholders like $(launchd_label) require user-supplied values.	4 / 5
Workflow Clarity	Six clearly sequenced steps with explicit safety checkpoints (idempotency, skip-if-exists, confirm-before-overwrite, marker-based replacement) and a one-at-a-time batch-ingest guardrail; the only gap is a missing explicit verify-the-wiki-works feedback loop beyond the final "tell the user to test" note.	4 / 5
Progressive Disclosure	Sections are well-organized (Steps 1-6 with 3a/3b/3c) and the single external reference (llm-wiki.md) is one level deep and clearly signaled, but detailed inline blocks like the ingest-discipline and URL-handling notes could live in separate reference files.	4 / 5
	Total	16 / 20 Passed

Description

78%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 that pairs a clear statement of purpose with explicit trigger phrases and a well-defined niche. The main weakness is specificity, since it describes a single high-level action rather than enumerating the concrete setup operations.

Suggestions

Expand the what-clause to name the concrete operations performed (create wiki/sources directories, generate a tailored container skill, wire a CLAUDE.md wiki section) to lift specificity from 3 toward 4-5.

Consider adding a synonym such as "notes" or "second brain" to the trigger list to broaden natural-term coverage.

Dimension	Reasoning	Score
Specificity	"Add a persistent wiki knowledge base to a NanoClaw group" names the domain and one concrete action but offers no broader enumeration of capabilities, matching the 1-2-concrete-actions anchor.	3 / 5
Completeness	It states clearly what the skill does and gives an explicit "Triggers on ..." clause with concrete trigger phrases, satisfying both what and when.	5 / 5
Trigger Term Quality	The trigger list ("add wiki", "wiki", "knowledge base", "llm wiki", "karpathy wiki") provides good natural-term coverage with synonyms, though a couple of plausible phrasings are absent.	4 / 5
Distinctiveness Conflict Risk	The NanoClaw/Karpathy wiki niche is distinct with specific triggers, but bare terms like "wiki" and "knowledge base" carry minor overlap risk with other knowledge-management skills.	4 / 5
	Total	16 / 20 Passed

Validation

100%

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 — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository: nanocoai/nanoclaw
Path: .claude/skills/add-karpathy-llm-wiki/SKILL.md
Commit: 858421a

Reviewed: 4 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.