Skip to content

Upstream MCP server

io.github.toolweather/upstreamtoolweatherstreamable-httpv0.2.0Scanned

Is this AI/dev API down? Is this model deprecated, what replaces it? Live status for ~40 APIs.

5 tools: 1 change things and 4 only read.

Tools: 0 delete, 0 irreversible, 1 write, 4 read.
  • Reads private data · 0
  • Reads text others write · 5
  • Can send data out · 0
  • 1 finding

Tools listed without sign-in. Remote posture

With Anlyon

1 tool here changes things. Govern it.

  • Anlyon lists the tools and adopts none.
  • Each tool you adopt gets approvals and receipts.
  • Your agent never holds the key.
Govern this server →

A tool your agent calls directly is not routed through Anlyon and is not governed by it. Anlyon Radar only lists and labels, and these labels are the same either way.

What the Upstream MCP server can do

The Upstream MCP server is published by the GitHub account toolweather. It has 5 tools. It can create or change data (1 tool: check_model_deprecation), but cannot delete data. 4 tools only read.

5 tools bring in text other people can write (check_api_status, check_model_deprecation, list_recent_api_changes and 2 more).

The fixed checks matched 1 finding (tool-shadowing). Each one comes with the text it matched.

Its remote endpoint lists its tools without sign-in, and its tools changed once since Anlyon Radar first saw them.

Last scanned Nov 21, 2026, version 0.2.0.

Tools

Grouped by what each tool does, most consequential first. Open one for its reasons and full description; descriptions are the publisher's own words.

5 tools

Changes things 1Claude Code: ask
  • check_model_deprecationCheck whether one LLM model id is active, deprecated, or retired, plus its shutdown date and replacement, from that provider's deprecation page (OpenAI, Anthropic, Google, and othe…Reads text others write1 finding

    Why it is in this group

    • Nothing in its name or description says what it does, so it is counted as a write.
    • The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.

    Findings

    • Warn

      The input schema at properties.id.description names a tool "get_model" that this server does not list.

      Matched text

      inputSchema.properties.id.description: "Deprecated alias of model, only for the get_model tool name. Same 1–120 character API id. Igno..."

      What this check looks for
    Description, as the server wrote it
    Check whether one LLM model id is active, deprecated, or retired, plus its shutdown date and replacement, from that provider's deprecation page (OpenAI, Anthropic, Google, and other tracked LLM vendors). Use before hard-coding or calling a model, or when a call fails with model-not-found. Not for API uptime (check_api_status) and not for a retirement feed (list_recent_api_changes with kind model_retirement, or get_digest). Returns summary, verdict (active|deprecated|retired|unknown), recommended_action (keep_using|plan_migration|switch_now|verify_with_provider), replacement, shutdown_date, evidence, and freshness; pages are re-read every few hours. An unlisted id is verdict unknown, not an error; longer than 120 characters returns 400 bad_model_id. Example: {"model":"gpt-4-0613"}. Free within the daily quota, then $0.003 via x402. Alias: get_model (param id).

    Annotations the server declares

    • title: Is this model deprecated?
    • readOnlyHint: true
    • destructiveHint: false
    • idempotentHint: true
    • openWorldHint: true
Only reads 4Show 4 read-only tools
  • check_api_statusCheck whether one AI or developer API provider (OpenAI, Anthropic, Gemini, Mistral, Groq, GitHub, Stripe, Cloudflare, Vercel, Supabase, Twilio, ~40 total) is up, degraded, or down …Reads text others writeNo findings

    Why it is in this group

    • The name "check_api_status" says "status": a word for reading.

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    Description, as the server wrote it
    Check whether one AI or developer API provider (OpenAI, Anthropic, Gemini, Mistral, Groq, GitHub, Stripe, Cloudflare, Vercel, Supabase, Twilio, ~40 total) is up, degraded, or down right now. Use before calling that provider, or right after 5xx/429/timeouts, to choose retry, wait, or a fallback. Not for a model id (check_model_deprecation), the filtered change feed (list_recent_api_changes), or the daily rollup (get_digest). Returns summary, verdict (up|degraded|partially down|down|unknown), recommended_action, incidents, affected_components, checked_at, and freshness; status is rechecked about every 4 minutes. An unknown provider id returns 404 unknown_provider. Example: {"provider":"openai"}. Free within the daily quota, then $0.003 via x402. Alias: get_status.

    Annotations the server declares

    • title: Is this API down?
    • readOnlyHint: true
    • destructiveHint: false
    • idempotentHint: true
    • openWorldHint: true
  • get_digestRead one published daily breaking-changes digest: AI/dev API status flips, model deprecations and retirements, pricing-page edits, and breaking MCP server changes.Reads text others writeNo findings

    Why it is in this group

    • The name "get_digest" says "get": a word for reading.

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    Description, as the server wrote it
    Read one published daily breaking-changes digest: AI/dev API status flips, model deprecations and retirements, pricing-page edits, and breaking MCP server changes. Use for a once-a-day catch-up. For a filtered on-demand API feed with no MCP section, use list_recent_api_changes; for live up/down, check_api_status; for one model id, check_model_deprecation. Always free (no quota, no x402). Published once at 12:00 UTC as a frozen window since the previous issue (usually ~24h); not updated later. Omit date for the latest; a missing YYYY-MM-DD returns 404 {error:no_issue}. One issue per date, no paging. Returns {date, title, summary, window_start, window_end, published_at, markdown, data, url, sponsor:null}. data holds counts, status, open_now, model_changes, shutting_down_30d, pricing, deprecation_page_edits, and mcp (breaking, non_breaking, down, back, auth_broken, auth_fixed). markdown caps model rows, breaking MCP rows, and each MCP event kind at 30, 30, and 20; data is the full window.

    Annotations the server declares

    • title: Daily digest
    • readOnlyHint: true
    • destructiveHint: false
    • idempotentHint: true
    • openWorldHint: true
  • list_providersList every AI and developer API provider Upstream tracks (~40) with id, name, category, current state, checked_at, and status page.Reads text others writeNo findings

    Why it is in this group

    • The name "list_providers" says "list": a word for reading.

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    Description, as the server wrote it
    List every AI and developer API provider Upstream tracks (~40) with id, name, category, current state, checked_at, and status page. Use to pick a provider id for check_api_status, or for one all-providers health pass. Not for one provider's incidents (check_api_status) and not for what changed (list_recent_api_changes or get_digest). No parameters. Always free; does not spend the daily quota. State is refreshed about every 4 minutes. Returns {count, providers:[{id, name, category, state, checked_at, status_page}]}; state is operational, degraded, partial_outage, major_outage, maintenance, or unknown.

    Annotations the server declares

    • title: List providers
    • readOnlyHint: true
    • destructiveHint: false
    • idempotentHint: true
    • openWorldHint: true
  • list_recent_api_changesList the on-demand feed of recent AI and developer API changes (status flips, deprecation-page edits, model retirements, pricing-page edits).Reads text others writeNo findings

    Why it is in this group

    • The name "list_recent_api_changes" says "list": a word for reading.

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    Description, as the server wrote it
    List the on-demand feed of recent AI and developer API changes (status flips, deprecation-page edits, model retirements, pricing-page edits). Use for a chosen window, one provider, or one kind, when explaining sudden breakage. Not for live up/down (check_api_status) and not for the once-a-day digest, which also covers MCP server breakage (get_digest). Returns summary, count, freshness, and changes[] with provider, kind, title, detail, source, and at, newest first. Status rows arrive within about 4 minutes; doc-page rows every few hours. A limit above 500 is clamped to 500; an unknown provider returns an empty list, not an error. Example: {"provider":"anthropic","kind":"model_retirement"}. Free within the daily quota, then $0.01 via x402. Alias: get_changes.

    Annotations the server declares

    • title: Recent API changes
    • readOnlyHint: true
    • destructiveHint: false
    • idempotentHint: true
    • openWorldHint: true

This is not a safety rating, and no finding is not a safety guarantee. The groups and findings come from fixed checks of the tool definitions the server returned on the date shown, not from its code or what it does when called. How the checks work.

Exfil surface

Which tools read private data, which read text someone else can write, and which can send data out. One of each is a path a prompt injection can use.

No private-data, untrusted-input, outbound path in this server alone: no tool is tagged private-data or outbound.

Reads private data0
Reads files, mail, messages, databases, secrets, calendars or repositories.
Reads text others write5
Brings in text someone else can write: web pages, inbound email, messages, issues and comments.
Can send data out0
Can carry data to people or systems: it sends, posts or publishes, or takes a recipient, channel or URL.

Other servers in the same agent add their own tools to these paths. Audit your whole setup to see them together.

Remote posture

What anyone on the internet sees at the remote endpoint: an anonymous initialize and tools/list, and the public OAuth metadata.

Tools listed without sign-in

Looked at Nov 20, 2026, 19:23 UTC, unchanged since Oct 22, 2026. No credentials were sent and no tool was called. What is looked at.

Reachable
yes
HTTP status of an anonymous initialize
200
tools/list served without credentials
yes
Tools that change things, listed without credentials
1
Protocol version
2025-06-18
WWW-Authenticate on 401
not observed
resource_metadata in WWW-Authenticate
not observed
RFC 9728 protected resource metadata
not observed
Authorization servers
not observed
Legacy HTTP+SSE transport
no
Dynamic client registration
not observed
Client ID metadata documents
not observed
  • WarnThe whole server

    The endpoint served tools/list without credentials, and the list has 1 tool that changes things (1 write). Whether a call also works without credentials was not tested: tools are never called.

    Matched text

    https://upstream.toolweather.workers.dev/mcp: tools/list returned 5 tools to an anonymous client, including check_model_deprecation (write)

    What this check looks for

Findings

Matches of fixed checks, with the text they matched.

1 warn
  • The input schema at properties.id.description names a tool "get_model" that this server does not list.

    Matched text

    inputSchema.properties.id.description: "Deprecated alias of model, only for the get_model tool name. Same 1–120 character API id. Igno..."

    What this check looks for

History

Changes to its tool definitions and to its remote endpoint, newest first.

  1. Tools3 changed.~3 changedView diff

Install

Pinned to version 0.2.0. Values in angle brackets are yours to fill in.

claude mcp add --transport http upstream https://upstream.toolweather.workers.dev/mcp
claude mcp add --transport http upstream https://upstream.toolweather.workers.dev/mcp

Run in a terminal.

Claude Code permissions

Allow the reads, ask before a change or a way out, deny the deletes. Paste into .claude/settings.json.

.claude/settings.json
{
  "permissions": {
    "allow": [
      "mcp__upstream__check_api_status",
      "mcp__upstream__list_recent_api_changes",
      "mcp__upstream__list_providers",
      "mcp__upstream__get_digest"
    ],
    "ask": [
      "mcp__upstream__check_model_deprecation"
    ],
    "deny": []
  }
}

Tools that only read are allowed, tools that change things ask first, tools that delete or cancel are denied. A read that can send data out (it takes a URL, say) asks first too. It assumes the server is installed as upstream. Review it before you use it.

Capability tags behind 5 rules
  • mcp__upstream__check_api_status untrusted-input
  • mcp__upstream__check_model_deprecation untrusted-input
  • mcp__upstream__list_recent_api_changes untrusted-input
  • mcp__upstream__list_providers untrusted-input
  • mcp__upstream__get_digest untrusted-input

Pin

Keep the decision: fail CI when the tools change, follow every change, or show a badge.

Pin in CI

Fail a build when the tools you reviewed are not the tools the server serves today.

curl -s "https://radar.anlyon.com/api/v1/servers/io.github.toolweather/upstream/verify?hash=7db0faedd4c82a4f0f02a6a0b73a26bf800a0c3c71cc23339bed48f30c5ba519" | jq -e '.changed == false'

Pinned hash 7db0faedd4c8, from Nov 21, 2026, 08:18 UTC.

Watch

Get told when its tools change: a new tool, a new description, a new class.

https://radar.anlyon.com/servers/io.github.toolweather/upstream/changes.xml

An Atom feed. Works in any feed reader or Slack's RSS app.

Badge

Anlyon Radar badge for io.github.toolweather/upstream
[![Anlyon Radar](https://radar.anlyon.com/badge/io.github.toolweather/upstream.svg)](https://radar.anlyon.com/servers/io.github.toolweather/upstream)

Source: MCP Registry · repository. Synced Oct 21, 2026, 18:50 UTC.

Upstream MCP server: questions and answers

Answered from the scan above, not from the publisher's description.

Is the Upstream MCP server safe?

No scan can promise that. What its definitions show: 5 tools, 1 of which can change, send or delete something, and 1 finding from fixed checks. Allow the reads, ask before each change, and deny the deletes; the permissions block on this page does that.

Can the Upstream MCP server delete data?

No tool is classed as destructive: going by their definitions, none of them deletes, cancels or revokes anything.

Can the Upstream MCP server do things that cannot be undone?

No tool is classed as irreversible: none of them sends, pays, publishes or merges, going by their definitions.

What tools does the Upstream MCP server have?

5 tools: 1 write, 4 read. check_model_deprecation, check_api_status, list_recent_api_changes, list_providers and get_digest.

Does the Upstream MCP server need sign-in or OAuth?

Its remote endpoint listed its tools to an anonymous client, without sign-in.

Does the Upstream MCP server send data to outside services?

None of its tools sends, posts or publishes, or takes a recipient, channel or URL.

Has the Upstream MCP server changed its tools?

Yes, once since Anlyon Radar started tracking it, most recently on Nov 7, 2026. Every change is on this page and in its Atom feed.