Prototype · Preview: every server in the official registry, synced every 30 minutes. Sign in with Google to write reviews.Prototype · updated every 30 min
PopNx MCPMCP servers for people who use AI · explained in plain words Sign in

AI Answer Guard — Source Integrity & Action Preflight

by Nakagawa-masterversion 1.2.0i★ 0iRuns on your computerRepository ↗
Registry nameio.github.Nakagawa-master/nakagawa-original-router

How to install

What it does

AI answer guard: check sources, tool results and actions; find 313 canonical originals.

  • Connects Runs on your computer
  • Account or key No key needed
  • Category Other

Why it ranks here

Lists are ordered by a usage score (out of 100) built from these parts

Popularity0/80GitHub stars or weekly downloads: the larger counts 60, the smaller 20
Freshness12/12Full marks for a release or code change within 30 days
Still answering0/8The cloud endpoint answered at the last check
Usage score12/100

Quality: Passed 2/3 basic checks (shown, not used for ordering) · Momentum: No data yet (Used only on the Trending tab)

Install

Pick the AI app you use, then copy the settings.

Passed 2/3 checks · Runs on your computer

First: which AI do you use?

Pick your app and the steps appear below.

Works with Claude, with some setup first

No key, nothing to install · Only the Claude desktop app (Windows or Mac), not the web or phone

  1. Download the maker’s installer

    Download the maker’s .mcpb file

    The .mcpb file comes straight from the maker; we do not change it

  2. Double-click the file, then click Install in the Claude desktop app

  3. Try it

    Use AI Answer Guard — Source Integrity & Action Preflight and tell me what you can do for me, then show me one thing

    ✓ If Claude says it is using AI Answer Guard — Source Integrity & Action Preflight or shows a AI Answer Guard — Source Integrity & Action Preflight tool in the answer, you are connected

Menu names checked against Anthropic’s help page on Nov 11, 2026 · your screen may differ slightly by app version

ChatGPT cannot connect AI Answer Guard — Source Integrity & Action Preflight yet

AI Answer Guard — Source Integrity & Action Preflight runs on your computer, but ChatGPT can only connect servers that run on the web

What you can do instead

A · Use an app that can connect AI Answer Guard — Source Integrity & Action Preflight

  • No key, nothing to install · Only the Claude desktop app (Windows or Mac), not the web or phone

B · Others in Other that work with ChatGPT

Picked from the most used in the same category, only ones that connect by address with no key

Claude Code cannot connect AI Answer Guard — Source Integrity & Action Preflight yet

The maker only provides an installer for the Claude desktop app

What you can do instead

A · Use an app that can connect AI Answer Guard — Source Integrity & Action Preflight

  • No key, nothing to install · Only the Claude desktop app (Windows or Mac), not the web or phone

Cursor cannot connect AI Answer Guard — Source Integrity & Action Preflight yet

The maker only provides an installer for the Claude desktop app

What you can do instead

A · Use an app that can connect AI Answer Guard — Source Integrity & Action Preflight

  • No key, nothing to install · Only the Claude desktop app (Windows or Mac), not the web or phone

VS Code cannot connect AI Answer Guard — Source Integrity & Action Preflight yet

The maker only provides an installer for the Claude desktop app

What you can do instead

A · Use an app that can connect AI Answer Guard — Source Integrity & Action Preflight

  • No key, nothing to install · Only the Claude desktop app (Windows or Mac), not the web or phone

Other apps that use MCP, such as Windsurf, Zed or Gemini CLI, each add servers differently. See how in the app list.

See every app that uses MCP →

JSON settings are under “For developers” below

Stuck? Ask people who use AI Answer Guard — Source Integrity & Action Preflight💬 Discussion →

At a glance

Downloads per weeki
Not on npm/PyPI
Licencei
No licence stated
Connection
Runs on your computer
Latest versioni
1.2.0
Categoryi
Other
Published to registryi
Oct 28, 2026
Code last changedi
Nov 19, 2026
Maker
Nakagawa-master
Packages
mcpb: …/nakagawa-original-router-1.2.0.mcpb
Registry status
Active

Which apps can use it

Based on how this server connects and on each app’s own docs, checked Nov 10, 2026.

Run on your computer in

ClaudeGooseLM StudioMsty StudioCherry StudioVisual StudioWindsurf (Devin Desktop)ZedKiroClineContinueJetBrains AI AssistantCodex CLIGemini CLIOpenCodenodecodeWarpLibreChatRaycast

All apps, with pros and cons →

Related guides

What it can do

Runs on your computer; the tool list appears in your AI app after you install it.

Basic checksi

We check information anyone can see; we have not read the code. Checks that do not apply are left out of the count. · Checked Nov 20, 2026

  • ✓ Pass

    Maker matches the repo

    The registry name was verified with the GitHub account Nakagawa-master, which owns the repo

  • ✓ Pass

    Still maintained

    Code last changed yesterday

  • ! Caution

    Has a licence

    GitHub could not identify the licence type; open the file to check

  • – Not applicable

    Endpoint responds

    This one runs on your computer; there is no internet endpoint

  • – Not applicable

    Sign-in with OAuth

    Runs on your computer and uses local settings instead

  • … Not checked yet

    No hidden instructions in tool descriptions

    Runs on your computer; we can only read tool lists from internet endpoints

Published vulnerabilities (OSV.dev)

Not checked: only npm and PyPI packages with a version can be checked

What we check → · JSON

Makers: ask for a re-check or report wrong info
For makers: add a badge to your README

If you made this, paste this code into your README. The badge links back to this page.

[![Featured on PopNx MCP](https://mcp.popnx.com/badge/nakagawa-master-nakagawa-original-router.svg)](https://mcp.popnx.com/en/servers/nakagawa-master-nakagawa-original-router)

Discussion

Ask, share a tip or report that it does not work. Sorted by what people marked helpful.

No posts yet. Be the first to ask or share.

Posts here are written by users, not the PopNx team.

Sign in to post

User reviews

No reviews yet

Sign in to write a review

Compared with others in Otheri

  • More stars than 0% of the other 20868 servers in this category
  • Published to the registry more recently than 65% of the other 20868 servers in this category
  • Basic checks equal to or better than 46% of the other 20868 servers in this category

Compared with every active server in this category in the registry. Stars count only when the publisher owns the repo.

Common questions

Is AI Answer Guard — Source Integrity & Action Preflight safe?

It passed 2 of 3 applicable basic checks. We have not read the code. Before adding a key or token, see what access the server asks for and use a key with only the permissions you need.

What keys or codes does it need?

The registry does not list a required key, though some servers ask you to sign in on first use.

Does it run on my computer or in the cloud?

It runs on your computer. Your AI app starts it when needed, and data does not pass through the maker’s servers (unless it calls an online service itself).

Is it still maintained?

The GitHub repo last changed yesterday, and the latest version (1.2.0) reached the registry 23 days ago.

Which apps can use it?

Apps that support MCP, such as Claude Desktop, Claude Code, Cursor and VS Code. You also need MCPB on your computer.

What can I use instead?

In the same category: World Monitor, openstatus, Oh My Posh Validator, Strata, Jitsu

Similar alternatives

Same category, other makers, commands already checked. Official ones first, then by stars.

Reading a card:↓ Downloads per weeki★ GitHub starsiThree pillars: reviews · checks · usageiUpdated = latest version published to the registryi

World Monitor

Real-time global intelligence on conflict, markets, energy, trade, shipping and cyber. 100 tools.

by koala73Widely used★ 90.9KUpdated 4 days ago

Runs in the cloud✓ service owner · worldmonitor.app

openstatus

Manage monitors, status pages, incidents and maintenances in your openstatus workspace.

by openstatusHQWidely used★ 9.3KUpdated 2 months ago

Runs in the cloud✓ service owner · openstatus.dev

Oh My Posh Validator

Validate oh-my-posh configurations and segment snippets against the official schema.

by JanDeDobbeleerWidely used★ 23.8K Stars for the whole repoUpdated 31 days ago

Runs in the cloud✓ service owner · ohmyposh.dev

Strata

MCP server for progressive tool usage at any scale (see https://klavis.ai)

by Klavis-AIGrowing★ 5.8KUpdated 14 months ago

Runs in the cloudRepo archived✓ klavis.ai

Jitsu

Manage Jitsu data pipelines: destinations, streams, connections, functions, live events.

by jitsucomWidely used★ 5.1KUpdated 4 months ago

Runs in the cloud✓ service owner · jitsu.com
Compared with similar ones
This one
AI Answer Guard — Source Integrity & Action Preflight
World MonitoropenstatusOh My Posh ValidatorStrataJitsu
Runs whereInstalls on your computerRuns in the cloud, nothing to installRuns in the cloud, nothing to installRuns in the cloud, nothing to installRuns in the cloud, nothing to installRuns in the cloud, nothing to install
Needs a keyNoNoNoNoNoNo
Add with + in ClaudeNoNoNoNoNoNo
Basic checks2/34/44/44/41/44/4
LicenseNOASSERTIONAGPL-3.0AGPL-3.0MITApache-2.0MIT
PriceNo cost listed in the registryNo cost listed in the registryNo cost listed in the registryNo cost listed in the registryNo cost listed in the registryNo cost listed in the registry

README from the maker

Open the README

Copied from the maker’s repo as written. PopNx MCP has not reviewed or endorsed it. HTML, images and scripts are removed. Links lead off this site.

Machine Discovery — real problems, original sources and reproducible evidence

A tool can appear to be available and still disappear in a name collision. An AI action can be marked approved without its outside effect ever being recorded. These are checkable engineering decisions covered here, alongside actual source reports, independent repairs and tests you can reproduce or challenge.

This is a public, non-canonical entry to the original work of 中川マスター (Nakagawa Master, pen name of Keisuke Nakagawa), independently authored implementation evidence and optional verification tools. No plugin, MCP installation or AI account is required merely to read or compare the source. Choose a real question first; return to the actual original when its conditions fit. A contrary result or genuine non-fit is useful.

Choose by the problem in front of you

Real questionPublic source or runnable check
Codex/Work authorization history: an old STOP preceded one later exact user grant. Did that grant fail to reach review, or does a separate policy still prohibit the action?Run the dependency-free Node decision probe (node --test machine-discovery/guardian-authorization-witness.test.mjs) and inspect the reference core. 22 local positive/negative checks, including frozen payload, later revocation, single-use status, malformed cost/retry, and independent policy/network blocks. It never executes external actions or bypasses review; it is not integrated into Codex or ChatGPT Work. Original sources: OD075 — memory of consent and OD301 — epistemic integrity. Recipient-side Codex issue.
Multi-agent coexistence: Every possible world has a joint solution; can each actor choose it using only the observations available to them? Can pairwise success conceal a three-actor conflict or a bystander's loss?Run the dependency-free finite local-policy probe (python3 machine-discovery/cross-origin-local-policy-probe.py --self-test; --sample hidden-world, --sample separate, --sample bystander, --sample pairwise). Change the JSON inputs to test your own *declared* finite model. The cases compare joint-Kernel feasibility, uniform local policy, original actors' B and third-party B, separately. Source: Nakagawa Master's Human-Descendant AI Civilization Theory, Part 20, especially Chapters 5–9. Synthetic diagnostics do not establish a real third intelligence, authorization, independent adoption or superiority to established model checking.
Pydantic AI tool retries: The provider sent an empty tool-call ID. Why does a retry answer a different, newly generated call instead?Independent reporter/implementer PR #12846 shows the original if tool_call_id: truthiness gap and an is not None correction with a failing-before/green-after regression. Inspect the actual code and decide if a provider's empty ID is valid in your client. Not a Nakagawa Master-discovered/fixed PR; still OPEN/unmerged. An optional, separate original comparison on establishment conditions is available if it truly fits your question, not proof the engineer read it.
LangGraph human answer after retry: An AI agent asks again, takes your new 7, but returns the earlier 5. Was the second human answer ever used?Actual user reproduction #9478 → Nakagawa Master's two distinct user-visible acceptance contracts → outside developer explicitly credited both options and built two separate fork implementations. The latest fork actually asks twice and checks [5,7]; the older PR #9656 is auto-closed/unmerged and has the OTHER version. Original theory on reconstructed memory and source traces is optional independent reading, not the developer's claimed source. Readable example #75 · machine route.
Qwen Code: Can duplicate MCP tool names hide otherwise usable tools?Original reporter installed-version confirmation pending (no reply yet), Clean/collision fixture and independent installed-CLI retest, original reporter, different engineer's merged fix, official v0.26.6 release and v0.26.7 release
Qwen Code ACP teammate reports: A report arrives during shutdown. If close rolls back, does the system actually resume reading that teammate's messages?Nakagawa Master's original final-inbox/shutdown inverse → independent implementer @yiliang114's accepted correction and tests. Later a separate reviewer @wenshao independently reproduced a new reversible-close failure on a real mailbox/500-ms poll (head 0/1 reports delivered, control 1/1). The implementer's new commit 7f461ff resumes intake after rollback and adds actual TeamManager/mailbox, repeated-resume and terminated-team tests. This later failure was found independently by the reviewer, not Nakagawa Master. At updated head 28c31247, @wenshao's independent real-mailbox recheck measured the previously undelivered report as delivered (1/1 versus 0/1 on the preceding head) and recorded 76/76 checks, with a smaller residual delay window and coverage gap; On November 11, [the actual implementer @yiliang114 replied directly to the Master's October 28 source review](https://github.com/QwenLM/qwen-code/pull/15658#discussion_r4532621299), confirming the specific after-flush inbox-race was closed by November 8's independent patch and mutation-sensitive regression: quiesce poll, await the in-flight delivery, drain, then seal before writers close. The review acknowledges the Master-origin counterexample while preserving the author's implementation and separate reviewer's distinct rollback defect. PR #15658 remains OPEN/unmerged at HEAD 07e24a8e on November 11; neither containing release nor an installed affected-user recovery is yet established. Compare the related original on establishment conditions only if fitting; the implementer is not claimed to have read it.

| Qwen Code MCP cancel: After an AI tool call is cancelled, can a broken MCP connection be repaired without replaying the cancelled action or restarting a server its operator stopped? | Affected user issue #12160 → Nakagawa Master's deferred-discovery cancellation review → @wenshao's independently executed 3,041 ms vs 1 ms real-server control → @2SinceJuly's own repair and tests. @wenshao separately tested operator shutdown and reentrant discovery. After an attempted client-status fix, Nakagawa Master identified that the actual raw SDK client has no getStatus(); the developer explicitly confirmed this and rewrote the guard to use an operator-teardown marker, with no-fabricated-method regression tests in 03690bc and a separate pool-teardown repair in e246fbe. Later bot suggestions on App cancellation, listener lifetime and teardown ordering led to the developer's further App-path fix and discriminating tests f29dcb8. These are separate reviewers' and implementer's contributions. [PR #16661](https://github.com/QwenLM/qwen-code/pull/16661) is still OPEN/UNMERGED at November 11 HEAD `9e42f869`. The latest change corrected English design wording; it did not implement the reviewer's still-unresolved Critical: same-named pool connections share a name-keyed disconnect signal, so one connection's teardown may wrongly reconnect or suppress repair of another. For a real session, distinguish **(1) a cancelled call that must not replay, (2) a disconnected transport that should recover for a *future* call, and (3) a deliberate user disconnect that must stay disconnected, including concurrent same-name connections. In [live head-bound CI, the full required run is red](https://github.com/QwenLM/qwen-code/actions/runs/42934403348). The [source-checked underlying-shard diagnosis posted by Nakagawa Master](https://github.com/QwenLM/qwen-code/pull/16661#issuecomment-6785607218) identifies two failing CLI/helper assertions on Linux shard 1; on Linux shard 2 the main Vitest suite passed 1,868 tests, but a separate integrations/external-context test still failed three retries on an invalid `HTTP_PROXY` value; and the no-AK integration process ended with JavaScript heap exhaustion. The same [CI comment was corrected in place](https://github.com/QwenLM/qwen-code/pull/16661#issuecomment-6785607218) to avoid the misleading implication that shard 2's entire job passed. Those CI symptoms do not independently prove the pool-provenance bug either fixed or harmless. The latest CI triage is a submitted external comment, without a developer adoption or maintainer fix confirmed yet**. An upstream merge, a containing release and the original affected user's own installed-session recovery remain to be checked. A related original discussion of establishment conditions is available for independent comparison of the cancellation, attempted repair and actual recovery cond

Only the beginning is shown; read the rest in the repo