Svelte LLMBLOCK
Svelte developer documentation as an MCP and in llms.txt format
Overview
From the repository's own README, as read at the audited commit. Badges and raw HTML are left out.
svelte-llm
LLM presets and MCP for Svelte 5 and SvelteKit. Visit the site at svelte-llm.stanislav.garden.
MCP Endpoint
This service provides an MCP (Model Context Protocol) endpoint for use with AI assistants:
- Streamable HTTP (Claude Desktop and most other clients):
https://svelte-llm.stanislav.garden/mcp/mcp - SSE (Older clients that don't support Streamable):
https://svelte-llm.stanislav.garden/mcp/sse
- Create a
.envfile with the content:
GITHUB_TOKEN= DB_URL=postgres://admin:admin@localhost:5432/db
- Create a Classic GitHub token. It must have
public_repopermissions. - Enter this in the .env file.
- Run
docker-compose up - Run:
nvm use npm i npm run dev
- Run database migrations: visit http://localhost:5173/api/migrate in your browser after starting the dev server.
You can also visit http://localhost:5173/admin to see all the "hidden" endpoints (default password = "secret")
Debug MCP
NODE_TLS_REJECT_UNAUTHORIZED=0 NODE_OPTIONS="--insecure-http-parser" npx @modelcontextprotocol/inspector
http://127.0.0.1:6274
You can also use GitHub Copilot in Agent mode to try the agent locally, see .vscode/mcp.json for info on how that works.
Misc
OG image from https://dynamic-og-image-generator.vercel.app/
c59a799c128cOBSERVED · 2026-10-06Connect
Built from this server's own package name, version and transport as found in its source — not copied from anyone's documentation, so it cannot drift against a page we do not control.
claude mcp add llmctx.com -- npx -y [email protected]
{
"mcpServers": {
"llmctx.com": {
"command": "npx",
"args": [
"-y",
"[email protected]"
]
}
}
}Exposed tools (2)
2 read · 0 write · 0 destructive.
| Tool | Risk | Description |
|---|---|---|
get_documentation | read | Retrieves full documentation content for Svelte 5 or SvelteKit sections. Supports flexible search by title (e.g., |
list_sections | read | Lists all available Svelte 5 and SvelteKit documentation sections in a structured format. Returns sections as a list of |
Trust audit
BLOCKgrade D · trust 69/100 Do not install this without reading the findings. The audit found something that could harm you or your machine.
| Layer | What it checks | Result |
|---|---|---|
| L0 | Provenance & inventory | PASS |
| L1 | Static analysis of the code | FAIL |
| L2 | Instruction surface (what it tells the agent) | PASS |
| L3 | Class-specific surface | WARN |
| L4 | Behavioural (sandbox) | SKIPPED |
What the source does
- Filesystem
- declared (1 observation(s))
- Network
- declared (5 observation(s))
- Shell
- none-observed
- Dependencies
- not all pinned
- Secrets in source
- found
Findings (12)
const DB_URL = 'postgres://admin:admin@localhost:5432/db'
connectionString: env.DB_URL || 'postgres://admin:admin@localhost:5432/db',
DB_URL=postgres://admin:admin@localhost:5432/db
sse,streamable-http
.cocoignore
.prettierignore
const migrationsPath = dev ? `${__dirname}/../../../../migrations` : '/app/migrations'http://127.0.0.1:6274
@anthropic-ai/sdk, @octokit/rest, @sveltejs/adapter-node, @sveltejs/kit, @sveltejs/vite-plugin-svelte, @testing-library/jest-dom, @testing-library/svelte, @types/eslint
- `CONTENT_SYNC_SECRET_KEY` - Secret key for content sync endpoint
mcp-publisher
- Visit `/admin` - Access admin panel (default password: "secret")
Gates applied: no_behavioural_pass.
c59a799c128cfull audit observations/trust-audit/mcp-server/khromov__svelte-llm.json · Report an issue / request a re-scanAudit history
Every audit this server has had. A grade with a past is a grade somebody is still checking.
| Date | Source | Verdict | Grade | Score | Change |
|---|---|---|---|---|---|
| 2026-10-06 | c59a799c128c | BLOCK | D | 69 | first audit |
Questions
What is the Svelte LLM MCP server?
Svelte developer documentation as an MCP and in llms.txt format
What tools does Svelte LLM expose?
2 in total: 2 read-only, 0 that write, and 0 that can delete or overwrite. Every one is listed on this page with its risk.
Is Svelte LLM safe to connect to an agent?
No — not without reading the findings first. The audit graded it D (69/100) and found 2 critical or high issues in the source. Each one is listed on this page with the file and line it is on.
What credentials does Svelte LLM need?
No credential environment variables were found in its source, so it appears to need none.
How does Svelte LLM run?
It speaks sse and streamable-http, so it runs as a service you connect to over the network. It is published on npm as llmctx.com at 0.0.1.
How current is this page?
The grade is for one exact copy of the source (c59a799c128c), read on 2026-10-06. The repository is watched and re-audited when it changes.