A local web server that displays a project wiki as a browsable website, with pages, research, chat, editing, AI assistance, and selectable themes.

Use it to start or stop the wiki website, browse pages in a browser, research topics, edit content in a split-pane view, and use the AI assistant.

It provides a visual way to read and work on wiki content instead of using only files or command-line queries.

wiki-serve — Skill for Claude Code by Oshayr/LLM-Wiki

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

Any agent

npx skills add Oshayr/LLM-Wiki --skill serve

Clone the repo

git clone --depth 1 https://github.com/Oshayr/LLM-Wiki

Made for: Claude Code.

Or install llm-wiki, the plugin that ships this one along with the rest of its 5 skills, 10 agents.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

README.md

[![agentmods](https://agentmods.dev/badge/skills/oshayr/llm-wiki/serve/github.svg)](https://agentmods.dev/skills/oshayr/llm-wiki/serve)

Your own site

<a href="https://agentmods.dev/skills/oshayr/llm-wiki/serve"><img src="https://agentmods.dev/badge/skills/oshayr/llm-wiki/serve/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

Your own site · 80×15

<a href="https://agentmods.dev/skills/oshayr/llm-wiki/serve"><img src="https://agentmods.dev/badge/skills/oshayr/llm-wiki/serve.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 48 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.

When invoked 551 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.

Origin original No closer match found in the catalogue.

Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

Model	Per session	Once invoked
Fable 5.1	$0.00048	$0.00551
Opus 5.5	$0.00019	$0.00220
Sonnet 5.5	$0.00010	$0.00110
Haiku 4.5	$0.00005	$0.00055

Measured 2mo ago against content hash edaa2563662b, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

wiki-serve scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2mo ago.

The scan reads SKILL.md. This mod also ships 11 executable files (scripts/__init__.py, scripts/chat_manager.py, scripts/markdown_renderer.py, …), listed below but not scanned — reading those needs a real analyzer, not pattern matching.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

skills/serve/SKILL.md · 62 lines

How it starts

The opening of the file, as written. The whole thing — 62 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Wiki Serve

Launch a local web server that presents the wiki as a browsable, Wikipedia-style website with live background research, integrated chat, and a split-pane editor with AI assist.

Resolve .wiki/ from plugin install scope. Auto-create if missing.

Arguments

/wiki-serve — start the server (default port 8420)
/wiki-serve stop — stop the running server

Startup

The FastAPI web server only runs when explicitly requested. There are two ways to start it:

Skill-Based Launch

Run the /wiki-serve skill from Claude Code (recommended):
```
/wiki-serve
```
The server automatically:
- Installs dependencies if needed
- Handles page indexing, research workers, and file watching
- Opens your browser to localhost:8420

Manual Launch

Dependencies are installed automatically by the plugin's SessionStart hook. To start the server manually from the command line:

python path/to/skills/serve/scripts/server.py --wiki-dir .wiki/ --port 8420

Replace path/to with the full path to your LLM-Wiki installation (typically .claude/plugins/llm-wiki).

If dependencies are missing, install manually: pip install -r requirements.txt

Features

Page rendering — markdown → HTML with wiki-link transformation, red-link detection
Live research — click any red link to trigger background research (stub-first: summary in 30s, full upgrade async)
Search — full-text search with autocomplete
Split-pane editor — markdown + live preview, markdown toolbar, AI assist bar
Page type templates — concept, brainstorming, idea, memory, status, rules, config, skill, plus custom templates from .wiki/templates/
Chat — wiki-aware AI assistant in bottom-right panel
Research queue — manage, cancel, reorder research tasks
Knowledge graph — interactive page relationship visualization
Three themes — light, dark, terminal
Export — HTML and markdown download

Stop

Kill the server process on port 8420. Either:

/wiki-serve stop
The server auto-stops when the session ends (daemon process)

Read the full file on GitHub · 62 lines

Files

What ships with it

29 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

2mo ago First seen · 62 lines · 48 tokens per session scan A edaa2563662b

Subscribe to this mod's changes

wiki-serve is a skill published in the GitHub repository Oshayr/LLM-Wiki (50 stars, last pushed 6mo ago), licensed MIT. It adds 48 tokens to every session and 551 once invoked, about $0.0002 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.

Browser automation Claude Code Research ai-tools autonomous-agent claude-code-plugin fastapi knowledge-base

llm-wiki

Build and maintain an LLM-curated knowledge base from papers, articles, transcripts, notes and project findings. Ingest sources into linked Markdown, answer questions with citations, lint or upgrade a wiki, and preserve useful synthesis across sessions and agents. Also capture verified task experience, consolidate…

praneybehl/llm-wiki-plugin · 143 tokens

webapp-testing

To test local web applications, write native Python Playwright scripts.

LiHongwei-cn/lihongwei-cn · 17 tokens

verify-feature

Auto-detect changed frontend routes, navigate them with Playwright, take screenshots, and check for console/network errors. Use after implementing a frontend feature to verify it works visually — auto-detects routes from git changes or accepts a route argument. Supports --responsive for mobile testing and…

makigjuro/cloudstack-ai-plugins · 65 tokens

screenshot

Capture a screenshot of a URL or route with one command. Use whenever the user wants to take a screenshot, capture a page, see what a route looks like, or visually verify a page. Supports mobile viewport with --mobile and unauthenticated mode with --no-auth.

makigjuro/cloudstack-ai-plugins · 58 tokens

mcp-tools

When and how to use the bundled and optional MCP tools — context7 library docs, Playwright browser automation, Mermaid diagram validation, and GitHub (gh CLI first). Apply when looking up library APIs, verifying frontend pages in a browser, validating diagrams, or doing GitHub operations.

makigjuro/cloudstack-ai-plugins · 61 tokens

smoke-test

Smoke test the running frontend and API by navigating key routes with Playwright, checking for console errors and broken pages. Use after starting the dev server or to verify the app works before creating a PR.

makigjuro/cloudstack-ai-plugins · 44 tokens