Sealmark
Automated re-test on 2026-11-20: not checked by hand, and not yet reviewed by the maintainers.

Results / paraplex-mcp

paraplex-mcp

version tested 0.3.0npmtested 2026-11-20Sealmark 1.0.1methodology 0.3-draftSTANDARD run30 tools

Sealmark's automated test of paraplex-mcp 0.3.0 on 2026-11-20 gave it grade B, its lowest test-area grade (protocol B). Only 4 of 30 tools could be called with default settings, so this grade covers the handshake and tool metadata only. Of its 6 checks, 3 passed, 1 failed and 2 not applicable. 3 more checks are advisory, and advisory checks never change a grade: 1 passed and 2 not applicable. The run used Sealmark 1.0.1 with the standard settings.

B

Grade B

The server's grade is its lowest test-area grade. How grading works

Only 4 of 30 tools could be called with default settings, so this grade covers the handshake and tool metadata only.

Grade by test area

B
Protocol3/4 points

Protocol checks

CheckVerdictWhat happened
protocol
Valid MCP handshake and tool schemas
pass
handshake ok, 30 well-formed tool(s)
Evidence for protocol: 1 line
  1. Probe initialize + tools/list
    Outcome server paraplex 0.3.0 answered; 30 tool(s)
    Decided from the MCP exchange (protocol)
tool-poisoningcritical
No tool-poisoning indicators in tool metadata
pass
scanned 30 tool definition(s) and serverInfo, including parameter text: nothing suspicious
robust-input
Survives malformed tool input
fail
server hung on malformed input
Evidence for robust-input: 3 lines
  1. Probe apply_template: required arguments missing
    Outcome no answer within 10s
    Decided from the MCP exchange (protocol) · tool apply_template
  2. Probe set_time: required arguments missing
    Outcome no answer within 10s
    Decided from the MCP exchange (protocol) · tool set_time
  3. Probe frame: every typed argument has the wrong type
    Outcome no answer within 10s
    Decided from the MCP exchange (protocol) · tool frame
secret-echocritical
No environment secret echoed in responses
pass
canary secrets absent from tools/list and 3 captured response(s), 0 of them successful result(s)
instructions-hygiene (advisory)
Server instructions name only tools that exist
n/a
the server sent no instructions
refusal-quality (advisory)
Refusals say what to do next
n/a
no refusal was recorded in this run, so there is nothing to judge

Effects checks

CheckVerdictWhat happened
annotation-honestycritical
Read-only tools change nothing
n/a
no tool declares readOnlyHint: true
openworld-honesty
Closed-world tools stay on the machine
n/a
no tool declares openWorldHint: false that could be given a harness-owned control argument
undeclared-writes (advisory)
Unannotated tools leave the sandbox unchanged
pass
render_variants, import_hyperframes, import_file, read_skill_file carry no annotations and changed nothing

Each check id links to what it tests and how to fix a failure. The same text is in sealmark explain <id>.

Under a check, Evidence shows what it probed (sent or inspected) and what it observed in this run: up to 6 lines per check, each cut at 240 characters, for checks that are not security checks. The text is shortened, and paths and anything that looks like a secret or token are replaced. Run sealmark run --format json yourself for every line in full.

By tool

Which checks called each tool with a valid call in this run, and which did not exercise it. A check can leave a tool out for several reasons: a limit Sealmark sets, a tool left out by design, input the run could not fill in, or a call that returned an error. The CLI report gives the reason. A verdict belongs to the whole check, not to one tool: a failed check that called three tools does not say which one failed. Security checks are left out of this table.

ToolCalled byNot exercised by
connectionNo check listed here called it with a valid call.
list_templatesNo check listed here called it with a valid call.
new_projectNo check listed here called it with a valid call.
apply_templateNo check listed here called it with a valid call.None
sceneNo check listed here called it with a valid call.
set_timeNo check listed here called it with a valid call.None
frameNo check listed here called it with a valid call.None
framesNo check listed here called it with a valid call.None
list_toolsNo check listed here called it with a valid call.None
run_toolNo check listed here called it with a valid call.None
export_pngNo check listed here called it with a valid call.None
exportNo check listed here called it with a valid call.
render_variantsNone
export_videoNo check listed here called it with a valid call.
import_hyperframesNone
import_fileNone
undoNo check listed here called it with a valid call.None
redoNo check listed here called it with a valid call.None
review_sheetNo check listed here called it with a valid call.None
review_artboardsNo check listed here called it with a valid call.None
list_projectsNo check listed here called it with a valid call.None
open_projectNo check listed here called it with a valid call.None
list_design_systemsNo check listed here called it with a valid call.None
apply_design_systemNo check listed here called it with a valid call.None
list_skillsNo check listed here called it with a valid call.None
load_skillNo check listed here called it with a valid call.
read_skill_file
write_skillNo check listed here called it with a valid call.
save_projectNo check listed here called it with a valid call.
scene_factsNo check listed here called it with a valid call.

What was not covered

Grade over time

Since the run on

Same as that run: the same grade in every test area, the same server version (0.3.0), the same Sealmark version (1.0.1).

  1. Bv0.3.0Sealmark 1.0.0
  2. Bv0.3.0Sealmark 1.0.1
  3. Bv0.3.0Sealmark 1.0.1

Package facts

License
MIT
Latest release
0.3.0, published 31 Oct 2026 (the version graded on this page)
Repository
Paraplex-AI/paraplex-mcp
Last push
31 Oct 2026
Open issues and pull requests
1

From npm and GitHub, checked 22 Nov 2026. Plain facts about the package; they play no part in the grade.

Add it to an MCP client

This starts version 0.3.0, the version graded on this page, with the start command from the test run. For a client that reads an mcpServers JSON file:

{
  "mcpServers": {
    "paraplex-mcp": {
      "command": "npx",
      "args": [
        "-y",
        "paraplex-mcp@0.3.0"
      ]
    }
  }
}

Or add it to Claude Code:

claude mcp add paraplex-mcp -- npx -y "paraplex-mcp@0.3.0"

The test run started the server with no API keys or other secrets, and with a cleaned environment. If the server needs some, its own documentation says how to pass them.

Run it yourself

This repeats the run with the same Sealmark version and the standard settings, starting the server the same way:

npx -y sealmark@1.0.1 run --start "npx -y paraplex-mcp@0.3.0"

To confirm your run used the standard settings, add --format json --out report.json to the command. In the report, config.standard should be true and config.hash should start with sha256:d13e7893a0ac, as they did in this run. If either differs, so did the settings, and the verdicts may too (STANDARD and CUSTOM runs).

To keep the setup, put the start command in sealmark.yaml as the server.

Grade alerts

One email when the grade of paraplex-mcp changes after a re-test. Nothing else, and every email has a link to stop them.

Or follow it in a feed reader: the Atom feed lists each move of the overall grade that came with a different version of this server, tested with the same Sealmark version as the graded run before.

Each of 613 verdicts was checked independently, and Sealmark was wrong on some: how reliable Sealmark is.

The issue we opened No reply yet

Sealmark opened a public issue on this project's repository on 17 Nov 2026 to share this grade. No reply yet. A reply is optional.

Read the issue on GitHub. State as of the last re-test run.

Is this wrong?

Reply on the GitHub issue or email disclosure@sealmark.dev and we will recheck it. Maintainers can ask for their server to be removed.

If we were wrong, we correct the page and add your reply. Security findings go to the maintainer privately first; see the disclosure policy.