Karpathy LLM Wiki
Astro-Han/karpathy-llm-wiki
A skill your agent uses when building or maintaining a personal LLM-powered knowledge base.
How to author a BrainDB custom profile — prompt add/replace fragments and an optional keyless ingestor — that shapes wiki naming/structure and feeds a custom ingestion source, with zero effect on…
$ npx skills add dimknaf/braindb --skill braindb-custom-profile -a claude-codeProject install by default; add -g for ~/.claude/skills/.
$ gh skill install dimknaf/braindb braindb-custom-profile --agent claude-codeProject scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).
$ git clone --depth 1 https://github.com/dimknaf/braindb.git skills-src && mkdir -p .claude/skills && cp -r skills-src/custom-profiles .claude/skills/braindb-custom-profile && rm -rf skills-srcUse ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.
Claude Code skills documentation · loads skills from .claude/skills/
Install the "braindb-custom-profile" agent skill from https://github.com/dimknaf/braindb/tree/main/custom-profiles into .claude/skills/braindb-custom-profile/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "braindb-custom-profile", then confirm the skill loads.Claude Code copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$skill-installer install https://github.com/dimknaf/braindb/tree/main/custom-profilesType this inside Codex. $skill-installer <name> installs a curated skill from openai/skills. The installer writes to $CODEX_HOME/skills (default ~/.codex/skills). Restart Codex if the skill does not show up.
$ npx skills add dimknaf/braindb --skill braindb-custom-profile -a codexProject install goes to .agents/skills/; add -g for ~/.codex/skills/.
$ gh skill install dimknaf/braindb braindb-custom-profile --agent codexProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/dimknaf/braindb.git skills-src && mkdir -p .agents/skills && cp -r skills-src/custom-profiles .agents/skills/braindb-custom-profile && rm -rf skills-srcUse ~/.agents/skills/ instead of .agents/skills for a personal install.
Codex skills documentation · loads skills from .agents/skills/
Install the "braindb-custom-profile" agent skill from https://github.com/dimknaf/braindb/tree/main/custom-profiles into .agents/skills/braindb-custom-profile/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "braindb-custom-profile", then confirm the skill loads.Codex copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add dimknaf/braindb --skill braindb-custom-profile -a cursorProject install goes to .agents/skills/; add -g for ~/.cursor/skills/.
$ gh skill install dimknaf/braindb braindb-custom-profile --agent cursorProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/dimknaf/braindb.git skills-src && mkdir -p .cursor/skills && cp -r skills-src/custom-profiles .cursor/skills/braindb-custom-profile && rm -rf skills-srcUse ~/.cursor/skills/ instead of .cursor/skills for a personal install.
Cursor skills documentation · loads skills from .cursor/skills/, .agents/skills/, .claude/skills/, .codex/skills/
Install the "braindb-custom-profile" agent skill from https://github.com/dimknaf/braindb/tree/main/custom-profiles into .cursor/skills/braindb-custom-profile/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "braindb-custom-profile", then confirm the skill loads.Cursor copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gemini skills install https://github.com/dimknaf/braindb.git --path custom-profiles--scope user (default) or --scope workspace; --path is the subfolder of the repo that holds the skill; --consent skips the security confirmation prompt.
$ npx skills add dimknaf/braindb --skill braindb-custom-profile -a gemini-cliProject install goes to .agents/skills/; add -g for ~/.gemini/skills/.
$ gh skill install dimknaf/braindb braindb-custom-profile --agent gemini-cliProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/dimknaf/braindb.git skills-src && mkdir -p .gemini/skills && cp -r skills-src/custom-profiles .gemini/skills/braindb-custom-profile && rm -rf skills-srcUse ~/.gemini/skills/ instead of .gemini/skills for a personal install, then run /skills reload.
Gemini CLI skills documentation · loads skills from .gemini/skills/, .agents/skills/
Install the "braindb-custom-profile" agent skill from https://github.com/dimknaf/braindb/tree/main/custom-profiles into .gemini/skills/braindb-custom-profile/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "braindb-custom-profile", then confirm the skill loads.Gemini CLI copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gh skill install dimknaf/braindb braindb-custom-profileInstalls for Copilot at project scope by default; add --scope user for a personal install. Preview a skill first with gh skill preview. Needs GitHub CLI 2.90.0 or later (public preview).
$ npx skills add dimknaf/braindb --skill braindb-custom-profile -a github-copilotProject install goes to .agents/skills/; add -g for ~/.copilot/skills/.
$ git clone --depth 1 https://github.com/dimknaf/braindb.git skills-src && mkdir -p .github/skills && cp -r skills-src/custom-profiles .github/skills/braindb-custom-profile && rm -rf skills-srcUse ~/.copilot/skills/ instead of .github/skills for a personal install. Commit .github/skills so cloud agent and code review can use it.
GitHub Copilot skills documentation · loads skills from .github/skills/, .claude/skills/, .agents/skills/
Install the "braindb-custom-profile" agent skill from https://github.com/dimknaf/braindb/tree/main/custom-profiles into .github/skills/braindb-custom-profile/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "braindb-custom-profile", then confirm the skill loads.GitHub Copilot copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add dimknaf/braindb --skill braindb-custom-profile -a opencodeOpenCode documents no install command of its own. Project install goes to .agents/skills/; add -g for ~/.config/opencode/skills/.
$ gh skill install dimknaf/braindb braindb-custom-profile --agent opencodeProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/dimknaf/braindb.git skills-src && mkdir -p .opencode/skills && cp -r skills-src/custom-profiles .opencode/skills/braindb-custom-profile && rm -rf skills-srcUse ~/.config/opencode/skills/ instead of .opencode/skills for a personal install.
OpenCode skills documentation · loads skills from .opencode/skills/, .claude/skills/, .agents/skills/
Install the "braindb-custom-profile" agent skill from https://github.com/dimknaf/braindb/tree/main/custom-profiles into .opencode/skills/braindb-custom-profile/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "braindb-custom-profile", then confirm the skill loads.OpenCode copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
braindb-custom-profileHow to author a BrainDB custom profile — prompt add/replace fragments and an optional keyless ingestor — that shapes wiki naming/structure and feeds a custom ingestion source, with zero effect on…
Braindb Custom Profile is an agent skill from dimknaf/braindb. How to author a BrainDB custom profile — prompt add/replace fragments and an optional keyless ingestor — that shapes wiki naming/structure and feeds a custom ingestion source, with zero effect on defaults when inactive.
Its SKILL.md is about 1.6k tokens, which your agent loads only when the skill is triggered. The skill folder holds 22 other files (for example `README.md`, `example/README.md` and `example/wiki_writer.add.md`).
It sits in Knowledge Management, covering LLM wikis. The repository describes itself as: An "LLM wiki" upgraded to a real database — typed entities, graph relations, HTTP API, and a built-in natural-language agent. The licence is Apache-2.0.
4 steps, taken from the first numbered list in SKILL.md.
Read from SKILL.md and the folder at commit 99cd121. It shows what the files ask for, not the result of running them.
Pre-approves these tools, so the agent can use them without asking each time:
ReadWriteEditBashFrom allowed-tools in the SKILL.md frontmatter.
Ships script files (Python, from the files we listed), which the agent can run.
Shell commands in SKILL.md call:
curldockerFrom the folder's file list and the shell code blocks in SKILL.md.
No URLs in SKILL.md. Its commands use curl and docker, which can reach the network depending on how they are called.
From URLs in SKILL.md, links to its own repository left out.
Names no API keys, tokens, secrets or passwords.
From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.
Braindb Custom Profile loads about 1.6k tokens when it runs. Until then it costs about 61 tokens; SKILL.md has 702 words of instructions outside code blocks.
Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.
The automated check noted patterns worth knowing about, such as sudo or a known installer.
ch:** `CUSTOM_PROFILE` in the repo-root `.env` names the active profile(s),`ingestor.py`, and an optional `.env`.├── .env # optional: the ingestor's own config/secrets- load config from a sibling `.env` (a tiny parser into `os.environ`, or `python-dotenv`);sibling `.env` so **no profile-specific variable ever appears in the publicame>/.env.example custom-profiles/<name>/.env` and fill it.2. In the repo-root `.env`: `CUSTOM_PROFILE=<name>` (or `a,b` for several).allowed-tools: Read, Write, Edit, BashAutomated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.
The full file from dimknaf/braindb at commit 99cd121, republished under its Apache-2.0 licence (© dimknaf). 702 words, ~1,576 tokens.
.claude/skills/braindb-custom-profile/SKILL.md (or your agent's skills folder). This skill also uses 18 other files; get the full folder from GitHub.A custom profile lets you (a) shape how the wiki maintainer/writer name and structure pages and (b) feed BrainDB a custom ingestion source — with zero effect on default behaviour when the profile is not active. One env switch turns a profile on; one self-contained folder holds everything it needs.
The committed custom-profiles/hackernews/ folder is a complete, keyless worked example
to copy from. custom-profiles/README.md is the short contract; this is the deep how-to.
CUSTOM_PROFILE in the repo-root .env names the active profile(s),
comma-separated. Unset ⇒ every prompt is byte-identical to the baked-in default and the
ingestor supervisor sleeps.custom-profiles/<name>/ holds the profile's prompt fragments, an optional
ingestor.py, and an optional .env.custom-profiles/<name>/
├── wiki_maintainer.add.md # appended to the maintainer prompt
├── wiki_maintainer.replace.md # OR replaces it entirely (advanced)
├── wiki_writer.add.md # appended to the writer prompt
├── wiki_writer.replace.md # OR replaces it entirely (advanced)
├── ingestor.py # optional: a standalone source feeder
├── .env # optional: the ingestor's own config/secrets
└── README.mdTargets are wiki_maintainer and wiki_writer. <target>.add.md is appended to the
base prompt; <target>.replace.md replaces it. With several active profiles, add
fragments are concatenated in CUSTOM_PROFILE order. All files are optional.
wiki_maintainer decides, per orphan entity: skip / create (with a proposed_name) /
attach / consolidate. Shape naming and dedup here (e.g. "name companies by official
name; same ticker = same company").wiki_writer authors the page body (free markdown). Shape structure here — the
sections, a labelled profile block, a dated chronicle.add over replace. add keeps the base prompt — and all the machine
contract it carries — intact and only layers your guidance on top. replace makes you
re-own that contract.[[ref:UUID]] inline citations — how a wiki links to its evidence;<!-- section:NAME --> markers — the section system (arbitrary NAMEs are allowed, so
you may add e.g. profile, background, current-developments);<!-- wiki:meta canonical_name=… keywords=a;b --> header — the only source of a
page's keywords.(unknown) — never invent. A structured field gets a value and its
[[ref:UUID]] only when a source supports it; otherwise write (unknown).**Tickers:** MSFT header line) so the extractor tags facts with it.A profile may ship ingestor.py: a standalone, long-running loop that feeds BrainDB. The
committed profile_runner sidecar launches each active profile's ingestor.py as an
isolated subprocess (restart-on-exit), and sleeps when no profile is active — so a broken
ingestor can never affect the api/watcher/wiki.
The simplest, least-coupled ingestor just writes files into data/sources/ and lets the
existing watcher do ingestion + fact-extraction — no DB or agent calls. Pattern:
.env (a tiny parser into os.environ, or python-dotenv);.state/ file), write one
<prefix>-<id>.md into data/sources/, with the entity made salient;data/sources/ingested/ (the verbatim text is stored in the DB at
ingest, so the file is only a carrier);Resolve data/sources/ from the script's own location so it works in the container and
standalone: Path(__file__).resolve().parents[1] / "data" / "sources". Load secrets from the
sibling .env so no profile-specific variable ever appears in the public
docker-compose.yml. Keep files small (title + a few fields).
cp custom-profiles/<name>/.env.example custom-profiles/<name>/.env and fill it..env: CUSTOM_PROFILE=<name> (or a,b for several).docker compose up -d — the api picks up the prompt shaping; profile_runner launches the
ingestor.ls data/sources/ # files dropped by the ingestor
curl -s "http://localhost:8000/api/v1/entities?entity_type=wiki&limit=20"
curl -s -X POST http://localhost:8000/api/v1/memory/context \
-H "Content-Type: application/json" -d '{"queries":["<an entity name>"]}'To deactivate: remove the CUSTOM_PROFILE line and docker compose up -d — defaults restored.
custom-profiles/hackernews/ is a complete, keyless profile (Hacker News → tech-entity
wikis). Copy its ingestor.py + wiki_*.add.md as a starting point, then run it with just
CUSTOM_PROFILE=hackernews — no key required.
custom-profiles/gdrive/ is a second example: a folder follower that ingests a Google Drive
folder incrementally — only new files and, on edits, only the changed sections (a .state
manifest + per-doc snapshots, each delta self-describing what part it is and where it belongs).
Copy it when your source is a watched folder of documents that change over time rather than a feed.
A note on privacy: keep real profiles that carry a domain prompt or an API key gitignored
(see .gitignore); only generic teaching profiles like example/ and hackernews/ are
committed.
© dimknaf, Apache-2.0. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file
SKILL.md and 18 other files in custom-profiles of dimknaf/braindb.
Open the folder on GitHubat commit 99cd121
Braindb Custom Profile next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.
| Skill | Stars | Used in | Tokens | Auto-check | Licence | Repo updated |
|---|---|---|---|---|---|---|
| Braindb Custom Profile this skilldimknaf/braindb | 111 | — | ~1.6k | Automated safety check: Notes | Apache-2.0 | |
| Karpathy LLM WikiAstro-Han/karpathy-llm-wiki | 2.4k | — | ~3.6k | Automated safety check: Pass | MIT | |
| LLM Wikilewislulu/llm-wiki-skill | 655 | — | ~3.7k | Automated safety check: Pass | None | |
| Wiki Builderrohitg00/pro-workflow | 2.9k | — | ~1k | Automated safety check: Pass | None | |
| Codex History IngestAr9av/obsidian-wiki | 3.5k | — | ~2.2k | Automated safety check: Notes | MIT | |
| Arkon Editnduckmink/arkon | 1.5k | — | ~1.6k | Automated safety check: Pass | Custom licence |
Astro-Han/karpathy-llm-wiki
A skill your agent uses when building or maintaining a personal LLM-powered knowledge base.
lewislulu/llm-wiki-skill
Build and maintain a Karpathy-style LLM knowledge base — a self-compiling Obsidian markdown wiki where an Agent ingests raw sources, compiles cross-linked concept/entity/summary pages, answers…
rohitg00/pro-workflow
Start, structure, and grow a persistent research wiki indexed in pro-workflow's SQLite knowledge base.
Ar9av/obsidian-wiki
Ingest Codex CLI conversation/session history into Obsidian as distilled knowledge.
nduckmink/arkon
Propose or directly apply edits to Arkon wiki pages, including proposing brand new pages.
dair-ai/dair-academy-plugins
Creates and maintains configurable research wikis: scaffold a folder, add sources, compile pages and indexes, and file query answers back.
dimknaf/braindb
Persistent memory across sessions via the BrainDB agent. An agent skill from dimknaf/braindb.
dimknaf/braindb
Memory recall and save. An agent skill from dimknaf/braindb.
Categories
How to author a BrainDB custom profile — prompt add/replace fragments and an optional keyless ingestor — that shapes wiki naming/structure and feeds a custom ingestion source, with zero effect on…. Braindb Custom Profile is an agent skill from dimknaf/braindb. How to author a BrainDB custom profile — prompt add/replace fragments and an optional keyless ingestor — that shapes wiki naming/structure and feeds a custom ingestion source, with zero effect on defaults when inactive.
Braindb Custom Profile fits situations like: tasks that involve LLM wikis.
Run `npx skills add dimknaf/braindb --skill braindb-custom-profile -a claude-code`. Or copy the skill folder (custom-profiles in dimknaf/braindb) into .claude/skills/braindb-custom-profile in your project. Claude Code loads it when a task matches its description.
Run `npx skills add dimknaf/braindb --skill braindb-custom-profile -a codex`. Or copy the skill folder (custom-profiles in dimknaf/braindb) into .agents/skills/braindb-custom-profile in your project. Codex loads it when a task matches its description.
Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add dimknaf/braindb --skill braindb-custom-profile -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/braindb-custom-profile, .gemini/skills/braindb-custom-profile, .github/skills/braindb-custom-profile and .opencode/skills/braindb-custom-profile in your project.
Going by SKILL.md and its folder, Braindb Custom Profile needs Python for the scripts in its folder and the command-line tools its instructions call (curl and docker). Our summary lists: Python 3; Docker. Its frontmatter pre-approves these tools: Read, Write, Edit, Bash.
SKILL.md contains no URLs. Its commands use curl and docker, which can reach the network depending on how they are called. This is read from the text; nothing was executed.
Our automated static check of SKILL.md found notes only (mentions a .env file; pre-approves every shell command (allowed-tools: bash)), nothing it rates as a warning. It is not a guarantee. Review the folder before installing.
Braindb Custom Profile is published under the Apache-2.0 licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.
About 1.6k tokens (SKILL.md is roughly 6.3k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.
Skills that share tags, products or a category with Braindb Custom Profile: Karpathy LLM Wiki (Astro-Han/karpathy-llm-wiki, 2.4k stars), LLM Wiki (lewislulu/llm-wiki-skill, 655 stars), Wiki Builder (rohitg00/pro-workflow, 2.9k stars) and Codex History Ingest (Ar9av/obsidian-wiki, 3.5k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.
dimknaf (a GitHub user) maintains it in dimknaf/braindb, which has 111 GitHub stars. The repository holds 3 skills in this directory. The repository was last updated on September 12, 2026.
Source: dimknaf/braindb on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.