Skills/PyPI/unaltraweb-mcp

unaltraweb-mcp@0.8.1

VERIFIED· install only

Publisher claims: "MCP control plane for unaltraweb website workspaces" Observed in sandbox: no network, touches nothing outside the project, writes no memory.

Registry PyPI (MCP server)Publisher dosquartsdedocs · publishing for 76 daysInstalls 388Last detonated 13 min ago
$claude mcp add unaltraweb-mcp -- uvx unaltraweb-mcp
WatchReport an issue

Coverage

Bundled scripts ran with no arguments; no agent task scenario for this bundle.
Tested
8 of 36 bundled scripts run with no arguments (the rest were not run: per-bundle cap), with decoy credentials in the environment · 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
UNALTRAWEB_WORKER_TOKENConfiguration, not credentials: BUNDLE_APP_CONFIG, MANUAL_PDF_IMAGE, MANUAL_RELEASE_SELECTOR, MCP_CONSUMER_WORKSPACE, UNALTRAWEB_DOCKER_ROOT, UNALTRAWEB_DOCKER_SOCKET, UNALTRAWEB_FACTORY_DIR, UNALTRAWEB_MCP_IMAGE, UNALTRAWEB_MCP_IMAGE_REFERENCE, UNALTRAWEB_PROJECT_USER, UNALTRAWEB_RUNTIME_SESSION
Where it talks to
no vendor host named in the codeAlso rubygems.org, doi.org, example.org — not the publisher's domain; check what for before connecting.
What we saw when it ran
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 rubygems.org, doi.org, example.org as well as its own vendor. The guard injects UNALTRAWEB_WORKER_TOKEN only into requests to the host you bind (and its subdomains) and refuses the rest.

$vet-skills run --net declared -- uvx unaltraweb-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 (2 scenarios; bundled scripts ran with no arguments; no agent task scenario for this bundle) on 2026-11-20 and scored it 85/100 (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 · 1 s · 2 scenarios · 0 deviations
t+CategoryObserved action
0.0sprocessno install scripts in bundle
0.0sprocessbundled script run in the sandbox with no arguments · not named in the instructions
0.1sprocessexited 2: artifact_handoff_v1.py: error: the following arguments are required: action, --workspace, --path
0.1sprocessbundled script run in the sandbox with no arguments · not named in the instructions
0.1sprocessexited 1: ImportError: attempted relative import with no known parent package
0.1sprocessbundled script run in the sandbox with no arguments · not named in the instructions
0.2sprocessexited 1: ImportError: attempted relative import with no known parent package
0.2sprocessbundled script run in the sandbox with no arguments · not named in the instructions
0.3sprocessexited 1: ImportError: attempted relative import with no known parent package
0.3sprocessbundled script run in the sandbox with no arguments · not named in the instructions
0.3sprocessexited 1: ImportError: attempted relative import with no known parent package
0.3sprocessbundled script run in the sandbox with no arguments · not named in the instructions
0.4sprocessbundled script run in the sandbox with no arguments · not named in the instructions
0.4sprocessexited 1: ImportError: attempted relative import with no known parent package
0.4sprocessbundled script run in the sandbox with no arguments · not named in the instructions
0.5sprocessexited 1: ImportError: attempted relative import with no known parent package
0.5sprocess8 of 36 bundled scripts run with no arguments (the rest were not run: per-bundle cap)
0.5sprocessscript backend runs setup only
0.5sprocessscript backend runs setup only
0.5sprocessno install scripts to re-run

Static findings

MEDIUM
Python subprocess execution detectedunaltraweb_mcp/manual_release.py:450 · Python subprocess execution detectedcompleted = subprocess.run(Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST05
MEDIUM
Python subprocess execution detectedunaltraweb_mcp/processes.py:43 · Python subprocess execution detectedprocess = subprocess.Popen(Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST05
MEDIUM
File type combined with upload actionunaltraweb_mcp/letter_bundle.py:27 · File type combined with upload action — agent may exfiltrate local data filesAST_ARGS = ["--standalone", "--to=json", "--output=/work/parsed.json", *INPUT_ARGS]Fix: Remove the call, or declare the host in frontmatter and explain in the README what is sent and why.
AST07
MEDIUM
File type combined with upload actionunaltraweb_mcp/site_tools.py:2910 · File type combined with upload action — agent may exfiltrate local data files…ed for web/PDF parity", "guidance": "Declare each referenced .vl.json or .vg.json source exactly once in .vegavisuals.yml, render with the companion vegavisuals factory, commit the output and lock, and never publish while visualization_che…Fix: Remove the call, or declare the host in frontmatter and explain in the README what is sent and why.
AST07
MEDIUM
Python subprocess execution detectedunaltraweb_mcp/bundler_runtime.py:206 · Python subprocess execution detectedos.execvpe(command[0], command, env)Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST05
MEDIUM
File type combined with upload actionunaltraweb_mcp/mcp_server.py:98 · File type combined with upload action — agent may exfiltrate local data files"For static Vega figures, reference a .vl.json or .vg.json source declared once in .vegavisuals.yml and use the companion vegavisuals tools to render and check its declared output."Fix: Remove the call, or declare the host in frontmatter and explain in the README what is sent and why.
AST07
MEDIUM
File type combined with upload actionunaltraweb_mcp/practice_pdf.py:553 · File type combined with upload action — agent may exfiltrate local data files"ast": "outputs/source.effective.json", "ast_sha256": artifacts["outputs/source.effective.json"]},Fix: Remove the call, or declare the host in frontmatter and explain in the README what is sent and why.
AST07
MEDIUM
File type combined with upload actionunaltraweb_mcp/practice_pdf.py:492 · File type combined with upload action — agent may exfiltrate local data files"ast": "outputs/source.effective.json", "ast_sha256": receipt["artifacts"]["outputs/source.effective.json"]}:Fix: Remove the call, or declare the host in frontmatter and explain in the README what is sent and why.
AST07
MEDIUM
Python subprocess execution detectedunaltraweb_mcp/manual_pdf_preview.py:124 · Python subprocess execution detectedcompleted = subprocess.run(Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST05
MEDIUM
File type combined with upload actionunaltraweb_mcp/manual_context.py:569 · File type combined with upload action — agent may exfiltrate local data files"manifest_sha256": digest(files["manual-context.json"]), "inputs": {key: digest(data) if data is not None else None for key, data in inputs.items()},Fix: Remove the call, or declare the host in frontmatter and explain in the README what is sent and why.
AST07
MEDIUM
Python subprocess execution detectedunaltraweb_mcp/docker_launcher.py:90 · Python subprocess execution detectedos.execv(command[0], command)Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST05
LOW
Dynamic module loadingunaltraweb_mcp/manual_context_sources.py:16 · Dynamic module loading (importlib.loader.exec_module via provenance)spec.loader.exec_module(loaded)Fix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.
AST05
LOW
Dynamic module loadingunaltraweb_mcp/manual_context_sources.py:15 · Dynamic module loadingloaded = importlib.util.module_from_spec(spec)Fix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.
AST05
MEDIUM
File type combined with upload actionunaltraweb_mcp-0.8.1.data/data/share/unaltraweb-launcher/mcp-factory.yml:79 · File type combined with upload action — agent may exfiltrate local data files- .vegavisuals.lock.json and outputs declared by .vegavisuals.ymlFix: Remove the call, or declare the host in frontmatter and explain in the README what is sent and why.
AST07
LOW
Dynamic module loadingunaltraweb_mcp/practice_pdf.py:96 · Dynamic module loadingmodule = importlib.util.module_from_spec(spec)Fix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.
AST05
LOW
Instructs the agent to modify its persistent memory or instruction filesunaltraweb_mcp/site_tools.py:1211 · Instructs the agent to modify its persistent memory or instruction files · informational: not an instruction file"Read AGENTS.md and edit the generated configuration and home page",Fix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.
AST01
LOW
Dynamic module loadingunaltraweb_mcp/practice_pdf.py:97 · Dynamic module loading (importlib.loader.exec_module via provenance)spec.loader.exec_module(module)Fix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.
AST05
MEDIUM
File type combined with upload actionunaltraweb_mcp/source_checks.py:245 · File type combined with upload action — agent may exfiltrate local data files"context/editorial-policy.json", ".unaltraweb/computations.lock.json")Fix: Remove the call, or declare the host in frontmatter and explain in the README what is sent and why.
AST07

Cost to the agent

Instructions loaded
≈50 tok
1 file · 52 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 unaltraweb-mcp@0.8.1 as of 2 min ago. This is the package itself, not its dependencies.