Skills/PyPI/safe-kaggle-mcp

safe-kaggle-mcp@0.4.2

VERIFIED· install only

Publisher claims: "A safety-first MCP server connecting Claude to Kaggle (competitions, datasets, kernels, models)." Observed in sandbox: no network, touches nothing outside the project, writes no memory.

Registry PyPI (MCP server)Publisher parkseokjune · publishing for 153 daysInstalls 162Last detonated 1 h ago
$claude mcp add safe-kaggle-mcp -- uvx safe-kaggle-mcp
WatchReport an issue

Coverage

Nothing executable in the bundle ran; the trace covers the install step only.
Tested
every file read, file written and host contacted by what ran
Not tested
use of the skill by an agent

What this means for the label: Verified here covers what ran. The server's tools were not exercised with a real credential, so nothing about their behavior is asserted; the credential summary below says what to do with that.

Before you give it your token

It will ask for
KAGGLE_API_TOKENConfiguration, not credentials: KAGGLE_CONFIG_DIR, KAGGLE_KEY, KAGGLE_MCP_ENABLE_DESTRUCTIVE, KAGGLE_MCP_ENABLE_PUBLISH, KAGGLE_MCP_SUBMISSION_CAP, KAGGLE_MCP_TRANSPORT, KAGGLE_MCP_WORKDIR, KAGGLE_USERNAME
Where it talks to
no vendor host named in the codeAlso kaggle.com — not the publisher's domain; check what for before connecting.
What the honeypot saw
no credential files read · no network
What we could not test
use of the skill by an agent

Connect it with a scoped credential, under the guard, and read the third-party hosts first. The code names kaggle.com as well as its own vendor. The guard injects KAGGLE_API_TOKEN only into requests to the host you bind (and its subdomains) and refuses the rest.

$vet-skills run --net declared -- uvx safe-kaggle-mcp

Never paste a credential into a chat, a ticket or a report, including ours. How the guard holds it.

Is [email protected] safe to install?

KillerSkills ran it in the KillerSkills honeypot sandbox (1 scenario; nothing executable in the bundle ran; the trace covers the install step only) on 2026-11-16 and scored it 90/100 (Verified). In what ran, nothing we could detect went wrong: no undeclared network egress, no credential reads, no persistent instruction writes. What it does with a real credential was not observed; the static findings and the publisher carry the rest of the score. Summary: no network, touches nothing outside the project, writes no memory.

Behavior trace

Honeypot run · 0 s · 1 scenarios · 0 deviations
t+CategoryObserved action
0.0sprocessno install scripts in bundle
0.0sprocessscript backend runs setup only
0.0sprocessscript backend runs setup only
0.0sprocessno install scripts to re-run

Static findings

MEDIUM
File permission modification commandsrc/kaggle_mcp/auth.py:59 · File permission modification command — agent changes file ownership/permissionslog.warning("Could not chmod %s: %s", p.name, e)Fix: Request the narrowest permission that works; replace wildcards with the exact paths or hosts.
AST05
MEDIUM
Process kill commandsrc/kaggle_mcp/eda.py:109 · Process kill command — agent can terminate processesexcept Exception as e: # noqa: BLE001 - one bad CSV shouldn't kill the restFix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.
AST05

Cost to the agent

Instructions loaded
≈62 tok
1 file · 40 in bundle

Tokens estimated at four characters each. Skills load their instruction text when triggered; the figure is what that costs.

Known vulnerabilities

No advisories in OSV for safe-kaggle-mcp@0.4.2 as of 40 min ago. This is the package itself, not its dependencies.