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

Results / @extremely-agency/mcp

@extremely-agency/mcp

version tested 0.1.1npmtested 2026-11-20Sealmark 1.0.1methodology 0.3-draftSTANDARD run31 tools

Sealmark's automated test of @extremely-agency/mcp 0.1.1 on 2026-11-20 gave it grade A, its lowest test-area grade (protocol A). None of the 31 tools could be called with default settings, so this grade covers the handshake and tool metadata only. Of its 6 checks, 4 passed and 2 inconclusive. 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.

A

Grade A

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

None of the 31 tools could be called with default settings, so this grade covers the handshake and tool metadata only.

Grade by test area

A
Protocol4/4 points

Protocol checks

CheckVerdictWhat happened
protocol
Valid MCP handshake and tool schemas
pass
handshake ok, 31 well-formed tool(s)
Evidence for protocol: 1 line
  1. Probe initialize + tools/list
    Outcome server extremely-agency 0.1.1 answered; 31 tool(s)
    Decided from the MCP exchange (protocol)
tool-poisoningcritical
No tool-poisoning indicators in tool metadata
pass
scanned 31 tool definition(s), the server instructions and serverInfo, including parameter text: nothing suspicious
robust-input
Survives malformed tool input
pass
119 invalid call(s) rejected cleanly
Evidence for robust-input: 1 line
  1. Probe 119 invalid call(s) across 24 tool(s)
    Outcome all rejected cleanly and the server stayed responsive
    Decided from the MCP exchange (protocol)
secret-echocritical
No environment secret echoed in responses
pass
canary secrets absent from tools/list and 120 captured response(s), 0 of them successful result(s)
instructions-hygiene (advisory)
Server instructions name only tools that exist
pass
the instructions name no tool the server lacks
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
inconclusive
every control call failed, so the read-only annotation could not be checked
openworld-honesty
Closed-world tools stay on the machine
inconclusive
every control call failed, so the closed-world annotation could not be checked
undeclared-writes (advisory)
Unannotated tools leave the sandbox unchanged
n/a
every tool carries annotations, nothing to check

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
list_agenciesNo check listed here called it with a valid call.
list_requestsNo check listed here called it with a valid call.None
answer_requestNo check listed here called it with a valid call.None
reply_to_requestNo check listed here called it with a valid call.None
give_taskNo check listed here called it with a valid call.None
list_agentsNo check listed here called it with a valid call.None
read_agentNo check listed here called it with a valid call.None
write_agent_fileNo check listed here called it with a valid call.None
list_modelsNo check listed here called it with a valid call.None
set_agent_modelsNo check listed here called it with a valid call.None
set_agentNo check listed here called it with a valid call.None
builder_guideNo check listed here called it with a valid call.None
check_buildNo check listed here called it with a valid call.None
create_agentNo check listed here called it with a valid call.None
delete_agentNo check listed here called it with a valid call.None
run_agentNo check listed here called it with a valid call.None
list_approvalsNo check listed here called it with a valid call.None
decide_approvalNo check listed here called it with a valid call.None
agency_todayNo check listed here called it with a valid call.None
agent_activityNo check listed here called it with a valid call.None
latest_reportNo check listed here called it with a valid call.None
list_integrationsNo check listed here called it with a valid call.None
read_integrationNo check listed here called it with a valid call.None
write_integrationNo check listed here called it with a valid call.None
write_pluginNo check listed here called it with a valid call.None
test_integrationNo check listed here called it with a valid call.
publish_integrationNo check listed here called it with a valid call.
test_eventNo check listed here called it with a valid call.
give_eventsNo check listed here called it with a valid call.
give_toolsNo check listed here called it with a valid call.
discard_integrationNo 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 (0.1.1), the same Sealmark version (1.0.1).

  1. Av0.1.1Sealmark 1.0.0
  2. Av0.1.1Sealmark 1.0.1
  3. Av0.1.1Sealmark 1.0.1

Package facts

License
AGPL-3.0-only
Latest release
0.1.1, published 5 Nov 2026 (the version graded on this page)
Repository
maxkxd/extremely-autonomous-agents
Last push
22 Nov 2026
Open issues and pull requests
49

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

{
  "mcpServers": {
    "extremely-agency-mcp": {
      "command": "npx",
      "args": [
        "-y",
        "@extremely-agency/mcp@0.1.1"
      ]
    }
  }
}

Or add it to Claude Code:

claude mcp add extremely-agency-mcp -- npx -y "@extremely-agency/mcp@0.1.1"

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 @extremely-agency/mcp@0.1.1"

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 @extremely-agency/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.

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.