vikunja-mcpBLOCK
MCP server for Vikunja. Lets Claude Code, Codex and Claude Desktop read and update your tasks, with a one-command setup CLI.
Overview
From the repository's own README, as read at the audited commit. Badges and raw HTML are left out.
An MCP server for Vikunja. It lets Claude Code, Codex, Claude Desktop and other AI agents read your board, create and update tasks, move cards between buckets, and leave comments.
npx -y @skylivellc/vikunja-mcp setup
Setup asks for your Vikunja URL and an API token, checks both, saves them, and adds the server to whichever of Claude Code, Codex and Claude Desktop you have installed. Restart the client and ask something like "What's open on my Vikunja board?"
Requires Node.js 22 or newer and Vikunja 2.6 or newer. CI runs the full tool suite against Vikunja 2.6 and 2.7.
Create an API token
In Vikunja, open Settings → API Tokens (https:///user/settings/api-tokens) and create a token with these permissions:
To keep agents read-only, grant only the read_* permissions and tasks_by_index and views_buckets. A tool that lacks a permission reports which one it needs instead of failing silently.
Tools
184e6ea8a9e6OBSERVED · 2026-10-14Connect
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. Replace the environment placeholders with a token scoped to the least it needs.
claude mcp add vikunja-mcp --env VIKUNJA_TOKEN=${VIKUNJA_TOKEN} -- npx -y @skylivellc/[email protected]{
"mcpServers": {
"vikunja-mcp": {
"command": "npx",
"args": [
"-y",
"@skylivellc/[email protected]"
],
"env": {
"VIKUNJA_TOKEN": "${VIKUNJA_TOKEN}"
}
}
}
}Exposed tools (10)
6 read · 4 write · 0 destructive.
| Tool | Risk | Description |
|---|---|---|
add_comment | write | |
create_task | write | |
get_task | read | |
list_buckets | read | |
list_comments | read | |
list_projects | read | |
list_tasks | read | |
list_views | read | |
move_task | write | |
update_task | write |
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 | PASS |
| L4 | Behavioural (sandbox) | SKIPPED |
What the source does
- Filesystem
- none-observed
- Network
- declared (1 observation(s))
- Shell
- declared (1 observation(s))
- Dependencies
- not all pinned
- Secrets in source
- found
Findings (9)
exec(file, args).then(
const password = "e2e-password-not-secret";
export const TOKEN = "tk_secret_test_token";
VIKUNJA_SERVICE_PUBLICURL: http://127.0.0.1:3456/
for _ in $(seq 1 30); do curl -sf http://127.0.0.1:3456/api/v1/info && exit 0; sleep 1; done
VIKUNJA_E2E_URL: http://127.0.0.1:3456
`VIKUNJA_E2E_URL=http://127.0.0.1:3456 pnpm test:e2e`. The test registers its own user and a token limited to the
@clack/prompts, @modelcontextprotocol/server, zod, @modelcontextprotocol/client, @types/node, typescript, vitest
In scope: the MCP server, the setup CLI, and how they store and send the token. Especially interesting:
Gates applied: no_behavioural_pass.
184e6ea8a9e6full audit observations/trust-audit/mcp-server/skylivellc__vikunja-mcp.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 |
|---|---|---|---|---|---|
| 2026-10-14 | 184e6ea8a9e6 | BLOCK | D | 69 | first audit |
Questions
What is the vikunja-mcp MCP server?
MCP server for Vikunja. Lets Claude Code, Codex and Claude Desktop read and update your tasks, with a one-command setup CLI.
What tools does vikunja-mcp expose?
10 in total: 6 read-only, 4 that write, and 0 that can delete or overwrite. Every one is listed on this page with its risk.
Is vikunja-mcp safe to connect to an agent?
No — not without reading the findings first. The audit graded it D (69/100) and found 1 critical or high issue in the source. Each one is listed on this page with the file and line it is on.
What credentials does vikunja-mcp need?
It reads VIKUNJA_TOKEN from the environment. Give it a token scoped to the least it needs — an agent that can be talked into calling a tool can be talked into calling it with your credentials.
How current is this page?
The grade is for one exact copy of the source (184e6ea8a9e6), read on 2026-10-14. The repository is watched and re-audited when it changes.