Not in the top 1,000 by GitHub stars

MehmetGoekce/llm-wiki

Build Karpathy's LLM Wiki with Claude Code. L1/L2 cache architecture. Logseq + Obsidian support.

History

Review weekly changes to SKILL.md before updating. Dates mark the weeks in which we observed each version; checks apply to the recorded commits.

View weekly changes

SKILL.md

Week of Nov 2, 2026 · +0 / −0 lines

Before Network or broad shell → After Network or broad shell

11538c119566 → 11538c119566

Compare on GitHub

Week of Oct 26, 2026 · Line counts unavailable

Before Not checked yet → After Network or broad shell

First observation → 11538c119566

No earlier version was observed for comparison.

Compare on GitHub

View on GitHubAdd to my skills

Verified commit 11538c1195666db24e25e23cc64bb1d70587aef9

git clone --filter=blob:none --no-checkout https://github.com/MehmetGoekce/llm-wiki.git .hotskills/mehmetgoekce-llm-wiki && git -C .hotskills/mehmetgoekce-llm-wiki checkout --detach 11538c1195666db24e25e23cc64bb1d70587aef9 # Copy each selected skill directory from .hotskills/mehmetgoekce-llm-wiki into .claude/skills/<skill-name>/
git clone --filter=blob:none --no-checkout https://github.com/MehmetGoekce/llm-wiki.git .hotskills/mehmetgoekce-llm-wiki && git -C .hotskills/mehmetgoekce-llm-wiki checkout --detach 11538c1195666db24e25e23cc64bb1d70587aef9 # Copy each selected skill directory from .hotskills/mehmetgoekce-llm-wiki into .agents/skills/<skill-name>/
162GitHub stars
9stars in 7 days
1skill in this repository
24commits in 30 days

Heat

5.1
  • Star growth, last 7 daysNot available
  • Star growth, last 24 hoursNot available
  • Rank movement, last 7 daysNot available
  • Commits, last 30 daysNot available
  • Accounts asking about it through MCP, last 7 daysNot available
  • Hot votes this weekNot available

Heat never changes the order by stars, and sponsorship never touches it. How heat works

Stars over time

Oct 25, 2026Nov 4, 2026162

30-day change not yet available: not enough history.

Individual skills in this repository

How it works

llm-wiki

When your agent uses it

Maintain a persistent LLM wiki in Logseq or Obsidian. Subcommands - setup (one-time, creates the wiki), ingest a source into wiki pages, query the wiki, prune cold pages or stale L1 rules, lint, status, import existing notes.

What it loads into context

Frontmatter

87 tokens

Always available to the agent

SKILL.md

8,184 tokens

When the skill is activated

Reference documents

70,554 tokens

Loaded on request

Token counts are estimates. Scripts run separately; they are not counted as reference context.

The playbook

  1. Arguments
  2. Workflow
  3. Configuration
  4. Tool-Specific Format Rules
  5. L1/L2 Boundary
  6. Workflow: ingest (Default)
  7. Workflow: query
  8. Workflow: prune (LRU-Demote, scheduled)
  9. Workflow: prune --l1 (L1-Verification)
  10. Workflow: lint
  11. Workflow: status
  12. Workflow: setup
  13. Workflow: import
  14. Hub-Index-Routing (format)
  15. Access-Log (format)
  16. State + History (format)
  17. L1 Frontmatter (format)

Scripts executed separately

  • scripts/init-wiki.sh
  • setup.sh

Files

  • .claude-plugin/icon.pngasset306,139 bytes
  • .claude-plugin/marketplace.jsonasset271 bytes
  • .claude-plugin/plugin.jsonasset915 bytes
  • .github/ISSUE_TEMPLATE/bug_report.mddoc630 bytes
  • .github/ISSUE_TEMPLATE/config.ymlasset204 bytes
  • .github/ISSUE_TEMPLATE/feature_request.mddoc586 bytes
  • .github/PULL_REQUEST_TEMPLATE.mddoc594 bytes
  • .github/SECURITY.mddoc1,140 bytes
  • .gitignoreasset41 bytes
  • CHANGELOG.mddoc17,031 bytes
  • CONTRIBUTING.mddoc3,191 bytes
  • LICENSEasset1,071 bytes
  • README.mddoc19,484 bytes
  • SECURITY.mddoc1,240 bytes
  • SKILL.mdcore32,194 bytes
  • SKILL.mdcore32,194 bytes
  • SKILL.mdcore32,194 bytes
  • SKILL.mdcore32,194 bytes
  • SKILL.mdcore32,194 bytes
  • SKILL.mdcore32,194 bytes
  • SKILL.mdcore32,194 bytes
  • SKILL.mdcore32,194 bytes
  • SKILL.mdcore32,194 bytes
  • config.example.ymlasset369 bytes
  • diagrams/ingest-flow.mmdasset382 bytes
  • diagrams/l1-l2-architecture.mmdasset702 bytes
  • diagrams/query-routing.mmdasset571 bytes
  • docs/faq.mddoc7,720 bytes
  • docs/graph-view.pngasset310,396 bytes
  • docs/l1-l2-architecture.mddoc11,624 bytes
  • docs/logseq-vs-obsidian.mddoc9,776 bytes
  • docs/schema-reference.mddoc24,827 bytes
  • docs/troubleshooting.mddoc9,434 bytes
  • examples/before-after.mddoc3,397 bytes
  • examples/l1-memory-examples.mddoc2,479 bytes
  • openspec/AGENTS.mddoc2,779 bytes
  • openspec/changes/archive/2026-10-11-add-plugin-packaging/design.mddoc7,638 bytes
  • openspec/changes/archive/2026-10-11-add-plugin-packaging/proposal.mddoc4,628 bytes
  • openspec/changes/archive/2026-10-11-add-plugin-packaging/specs/config.mddoc1,680 bytes
  • openspec/changes/archive/2026-10-11-add-plugin-packaging/specs/plugin.mddoc6,102 bytes
  • openspec/changes/archive/2026-10-11-add-plugin-packaging/specs/setup.mddoc3,468 bytes
  • openspec/changes/archive/2026-10-11-add-plugin-packaging/tasks.mddoc3,976 bytes
  • openspec/project.mddoc2,839 bytes
  • openspec/specs/config.mddoc8,953 bytes
  • openspec/specs/ingest.mddoc12,949 bytes
  • openspec/specs/l1-l2-routing.mddoc13,021 bytes
  • openspec/specs/lint.mddoc16,523 bytes
  • openspec/specs/plugin.mddoc7,252 bytes
  • openspec/specs/prune.mddoc17,945 bytes
  • openspec/specs/query.mddoc12,680 bytes
  • openspec/specs/schema.mddoc12,458 bytes
  • openspec/specs/setup.mddoc13,201 bytes
  • scripts/init-wiki.shruns6,566 bytes
  • setup.shruns5,420 bytes
  • templates/logseq/Access-Log.mddoc374 bytes
  • templates/logseq/Dashboard.mddoc278 bytes
  • templates/logseq/Hub.mddoc478 bytes
  • templates/logseq/Schema.mddoc4,634 bytes
  • templates/obsidian/Access-Log.mddoc368 bytes
  • templates/obsidian/Dashboard.mddoc274 bytes
  • templates/obsidian/Hub.mddoc458 bytes
  • templates/obsidian/Schema.mddoc4,382 bytes
  • wiki.mddoc254 bytes
Read SKILL.md

# /llm-wiki - LLM Wiki

Persistent knowledge management powered by Claude Code. Maintains a structured wiki in Logseq or Obsidian using the L1/L2 cache architecture.

**Architecture: L1/L2 Cache Model**
- L1 = Claude Memory (auto-loaded): Rules, gotchas, identity, credentials
- L2 = Wiki (on-demand): Projects, workflows, research, deep knowledge

**Two cache mechanisms keep L2 precise as it grows:**
- **Hub-Index-Routing** — L1's auto-loaded memory index has no L2 counterpart by default. Each hub page
  carries an `### Index` of routing lines (one per child: `[[link]] -- description #tags`). `query`
  becomes two-stage: read the cheap hub indexes -> pick the 3 most relevant pages by description ->
  read only those full pages. This is the wiki's "page table / TLB" — no more grep-over-everything.
- **LRU-Demote** — `query` logs every page hit; `prune` evicts cold pages (no access in N months) from
View the complete source on GitHub

Skills included

  • llm-wiki

Repository facts

License
MIT
Created
Apr 8, 2026
Last pushed
Oct 11, 2026
Forks
29
Open issues
1
Project activity
Active

Before you install

Network or broad shell

Automated signals, not a guarantee

Checked Nov 6, 2026

Not fully analyzed: some files were too long, unreadable or missing. This never lowers the level.

Show 8 notes
  • SKILL.md:1 Analysis incomplete: Referenced executable target unavailable; not analyzed
  • SKILL.md:1 Analysis incomplete: Referenced executable target unavailable; not analyzed
  • SKILL.md:1 Analysis incomplete: Referenced executable target unavailable; not analyzed
  • SKILL.md:1 Analysis incomplete: Referenced executable target unavailable; not analyzed
  • SKILL.md:1 Analysis incomplete: Referenced executable target unavailable; not analyzed
  • SKILL.md:1 Analysis incomplete: Referenced executable target unavailable; not analyzed
  • SKILL.md:1 Analysis incomplete: Referenced executable target unavailable; not analyzed
  • SKILL.md:1 Analysis incomplete: Referenced executable target unavailable; not analyzed
Show all 65 signals by file

.claude-plugin/plugin.json

  • Network accessline 4
    "version": "1.7.1",   "description": "Persistent L1/L2 knowledge base in Logseq or Obsidian: ingest, query, prune, lint, status",   "author": { "name": "Mehmet Goekce", "url": "https://github.com/MehmetGoekce" },
  • Network accessline 5
    "description": "Persistent L1/L2 knowledge base in Logseq or Obsidian: ingest, query, prune, lint, status",   "author": { "name": "Mehmet Goekce", "url": "https://github.com/MehmetGoekce" },   "homepage": "https://github.com/MehmetGoekce/ll
  • Network accessline 6
    "author": { "name": "Mehmet Goekce", "url": "https://github.com/MehmetGoekce" },
  • Network accessline 7
    "homepage": "https://github.com/MehmetGoekce/llm-wiki",
  • Network accessline 8
    "repository": "https://github.com/MehmetGoekce/llm-wiki",
  • Network accessline 9
    "documentationUrl": "https://github.com/MehmetGoekce/llm-wiki#readme",
  • Network accessline 10
    "supportUrl": "https://github.com/MehmetGoekce/llm-wiki/issues",
  • Network accessline 11
    "privacyPolicyUrl": "https://github.com/MehmetGoekce/llm-wiki#privacy",

.github/ISSUE_TEMPLATE/config.yml

  • Network accessline 2
    contact_links:   - name: Question or Discussion     url: https://github.com/MehmetGoekce/llm-wiki/issues
  • Network accessline 3
    - name: Question or Discussion     url: https://github.com/MehmetGoekce/llm-wiki/issues     about: For questions, open a blank issue. No question is too small.
  • Network accessline 4
    url: https://github.com/MehmetGoekce/llm-wiki/issues

CHANGELOG.md

  • Network accessline 5
    Format based on [Keep a Changelog](https://keepachangelog.com/en/1.1.0/).
  • Network accessline 6
    This project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
  • Network accessline 321
    [1.5.1]: https://github.com/MehmetGoekce/llm-wiki/compare/v1.5.0...v1.5.1
  • Network accessline 322
    [1.5.0]: https://github.com/MehmetGoekce/llm-wiki/compare/v1.4.0...v1.5.0
  • Network accessline 323
    [1.4.0]: https://github.com/MehmetGoekce/llm-wiki/compare/v1.3.0...v1.4.0
  • Network accessline 324
    [1.3.0]: https://github.com/MehmetGoekce/llm-wiki/compare/v1.2.0...v1.3.0
  • Network accessline 325
    [1.2.0]: https://github.com/MehmetGoekce/llm-wiki/compare/v1.1.1...v1.2.0
  • Network accessline 326
    [1.1.1]: https://github.com/MehmetGoekce/llm-wiki/compare/v1.1.0...v1.1.1
  • Network accessline 327
    [1.1.0]: https://github.com/MehmetGoekce/llm-wiki/compare/v1.0.0...v1.1.0
  • Network accessline 328
    [1.0.0]: https://github.com/MehmetGoekce/llm-wiki/releases/tag/v1.0.0

CONTRIBUTING.md

  • Network accessline 15
    1. Fork the repo 2. Clone your fork: `git clone https://github.com/YOUR_USERNAME/llm-wiki.git`
  • Network accessline 16
    2. Clone your fork: `git clone https://github.com/YOUR_USERNAME/llm-wiki.git`
  • Dangerous shell operationline 41
    rm -rf /tmp/test-logseq /tmp/test-obsidian

README.md

  • Network accessline 3
    [![License: MIT](https://img.shields.io/github/license/MehmetGoekce/llm-wiki)](LICENSE)
  • Network accessline 4
    [![Release](https://img.shields.io/github/v/release/MehmetGoekce/llm-wiki)](https://github.com/MehmetGoekce/llm-wiki/releases)
  • Network accessline 5
    [![Stars](https://img.shields.io/github/stars/MehmetGoekce/llm-wiki?style=social)](https://github.com/MehmetGoekce/llm-wiki/stargazers)
  • Network accessline 6
    [![Top Language](https://img.shields.io/github/languages/top/MehmetGoekce/llm-wiki)](https://github.com/MehmetGoekce/llm-wiki)
  • Network accessline 7
    [![Last Commit](https://img.shields.io/github/last-commit/MehmetGoekce/llm-wiki)](https://github.com/MehmetGoekce/llm-wiki/commits)
  • Network accessline 9
    Build [Karpathy's LLM Wiki](https://gist.github.com/karpathy/442a6bf555914893e9891c11519de94f) with Claude Code. Two-layer cache architecture (L1/L2). Supports Logseq and Obsidian.
  • Network accessline 93
    git clone https://github.com/MehmetGoekce/llm-wiki.git
  • Network accessline 366
    - Inspired by [Andrej Karpathy's LLM Wiki gist](https://gist.github.com/karpathy/442a6bf555914893e9891c11519de94f)
  • Network accessline 367
    - Built with [Claude Code](https://claude.ai/code)
  • Network accessline 368
    - Works with [Logseq](https://logseq.com/) and [Obsidian](https://obsidian.md/)

SECURITY.md

  • Network accessline 11
    Report it privately through GitHub: [Report a vulnerability](https://github.com/MehmetGoekce/llm-wiki/security/advisories/new). The report is visible only to the maintainer.

docs/faq.md

  • Network accessline 15
    Yes. llm-wiki assumes you are using [Claude Code](https://claude.ai/code), which requires an Anthropic account and a paid plan (Pro or higher). All `/llm-wiki` commands run through Claude Code — there are no external API calls, no vector da
  • Network accessline 104
    [GitHub Issues](https://github.com/MehmetGoekce/llm-wiki/issues) — there are templates for bug reports and feature requests. Security issues: see [SECURITY.md](../SECURITY.md).

docs/schema-reference.md

  • Network accessline 297
    | Wiki internal | `[[Wiki/Tech/Strapi]]` | Referencing another wiki page | | Tag | `#strapi` | Lightweight categorization | | External | `[Strapi Docs](https://docs.strapi.io)` | URLs outside the wiki |
  • Network accessline 299
    | External | `[Strapi Docs](https://docs.strapi.io)` | URLs outside the wiki |

docs/troubleshooting.md

  • Network accessline 3
    Common issues and fixes. If your problem is not listed here, open a [GitHub issue](https://github.com/MehmetGoekce/llm-wiki/issues/new/choose).
  • Dangerous shell operationline 14
    - **Ubuntu/Debian:** `sudo apt install python3`
  • Network accessline 14
    - **Ubuntu/Debian:** `sudo apt install python3` - **Fedora:** `sudo dnf install python3` - **Windows:** Install from [python.org](https://www.python.org/downloads/) and check "Add python.exe to PATH" during install.
  • Dangerous shell operationline 15
    - **Fedora:** `sudo dnf install python3`
  • Network accessline 16
    - **Windows:** Install from [python.org](https://www.python.org/downloads/) and check "Add python.exe to PATH" during install.
  • Dangerous shell operationline 27
    - **Ubuntu/Debian:** `sudo apt install git`
  • Network accessline 27
    - **Ubuntu/Debian:** `sudo apt install git` - **Fedora:** `sudo dnf install git` - **Windows:** Install [Git for Windows](https://git-scm.com/download/win).
  • Dangerous shell operationline 28
    - **Fedora:** `sudo dnf install git`
  • Network accessline 29
    - **Windows:** Install [Git for Windows](https://git-scm.com/download/win).
  • Dangerous shell operationline 42
    - Check the path you entered is writable: `touch ~/Documents/Logseq/test.txt && rm ~/Documents/Logseq/test.txt`.

examples/l1-memory-examples.md

  • Access to secretsline 77
    [Credential-related source omitted]

openspec/changes/archive/2026-10-11-add-plugin-packaging/design.md

  • Network accessline 36
    "version": "1.6.0",   "description": "Persistent L1/L2 knowledge base in Logseq or Obsidian: ingest, query, prune, lint, status",   "author": { "name": "Mehmet Goekce", "url": "https://github.com/MehmetGoekce" },
  • Network accessline 37
    "description": "Persistent L1/L2 knowledge base in Logseq or Obsidian: ingest, query, prune, lint, status",   "author": { "name": "Mehmet Goekce", "url": "https://github.com/MehmetGoekce" },   "homepage": "https://github.com/MehmetGoekce/ll
  • Network accessline 38
    "author": { "name": "Mehmet Goekce", "url": "https://github.com/MehmetGoekce" },
  • Network accessline 39
    "homepage": "https://github.com/MehmetGoekce/llm-wiki",
  • Network accessline 40
    "repository": "https://github.com/MehmetGoekce/llm-wiki",

openspec/specs/ingest.md

  • Network accessline 124
    AND the wiki has a Schema page and a Wiki/Tech hub page AND no page exists for "Redis" WHEN the user runs /llm-wiki ingest "https://redis.io/docs/about/"
  • Network accessline 125
    AND no page exists for "Redis" WHEN the user runs /llm-wiki ingest "https://redis.io/docs/about/" THEN the system SHALL create a new page Wiki___Tech___Redis.md
  • Network accessline 126
    WHEN the user runs /llm-wiki ingest "https://redis.io/docs/about/"

scripts/init-wiki.sh

  • Executable scriptline 1
    #!/bin/bash
  • Dangerous shell operationline 97
    trap 'rm -f "$CREATED_LIST"' EXIT
  • Network accessline 204
    Generated by https://github.com/MehmetGoekce/llm-wiki") > /dev/null 2>&1; then

setup.sh

  • Executable scriptline 1
    #!/bin/bash
  • Dangerous shell operationline 114
    rm "$CONFIG_FILE"
  • Network accessline 166
    echo -e "  4. For automatic updates, install the plugin: ${CYAN}claude plugin marketplace add MehmetGoekce/llm-wiki${NC}" echo "" echo -e "Documentation: ${CYAN}https://github.com/MehmetGoekce/llm-wiki${NC}"
  • Network accessline 168
    echo -e "Documentation: ${CYAN}https://github.com/MehmetGoekce/llm-wiki${NC}"

How the signals fit together

Facts from a static read of the files at the pinned commit. They describe what the code does, not why.

Some files could not be read; the facts below still hold.

Signal confidence: 54 facts

and 4 more

Works with

Agent compatibility, from the files in this repository
AgentStatusEvidenceInstall to
Claude CodeNativeManifest .claude-plugin/marketplace.json.claude/skills/<skill-name>/
CodexNativeManifest openspec/AGENTS.md.agents/skills/<skill-name>/
CursorCompatibleStandard SKILL.md SKILL.md.cursor/skills/<skill-name>/
Gemini CLICompatibleStandard SKILL.md SKILL.md.gemini/skills/<skill-name>/
GitHub CopilotCompatibleStandard SKILL.md SKILL.md.github/skills/<skill-name>/
OpenCodeCompatibleStandard SKILL.md SKILL.md.opencode/skills/<skill-name>/
  • Native The repository ships a manifest, folder or frontmatter field for this agent.
  • Compatible A standard SKILL.md that this agent can read.
  • Unknown No evidence either way.

Derived automatically from paths, manifests and frontmatter. Not tested on each agent.