Fetch and follow the instructions at https://qamcp.dev/setup.md to test my MCP server.Your AI finds your server in your project, tests it with QAmcp, explains what's wrong in plain words, fixes it with your OK, and tests again. Read the instructions
Directory / Stable News
Stable News
by stablenews.ai · https://stablenews.ai/mcp/news
One news edition a day, ranked by a published rule, in our own words with links to the outlets.
13 of 13 checks · 8 Nov 2026
At a glance
- Runs where
- On their server. What you send to it goes to them.
- What you need
- Nothing: it answered without a login
- Tools
- 13 tools
- Its own safety labels
- 9 read-only, 1 labeled as deleting or destroying, 3 labeled as changing things
- Memory it takes
- About 3,800 tokens of your AI's memory, just for its tool descriptions
- Largest answer
- About 15.2 KB in our test
- Speed
- Average answer in our test: 470 ms
Tokens are an estimate: the characters in its tool names, descriptions and inputs, divided by 4. The exact number depends on your AI model. Safety labels are what the server says about itself. From our latest test, 8 Nov 2026.
Is Stable News MCP server safe?
- None of the hidden-instruction wordings we know were found in its 13 tool descriptions.
- Its error messages showed no secrets or internal details.
- 9 of its 13 tools are labeled read-only; the others may change or send things.
- Overall it scored 94/100 on 8 Nov 2026.
A test can only see what the server shows from outside, not what it does with your data behind the scenes. Connect servers from makers you trust.
What this test does not check
- Its source code and packages.
- What a tool quietly does behind the scenes, such as writing a file or calling another service.
- Instructions hidden inside live tool answers. We only look for those in tool descriptions.
- Hidden instructions worded in a way our list doesn't know yet.
- Tools that could change something. Those are read, never run.
What QAmcp can and can't see · How to tell if an MCP server is safe · Hidden instructions · Prompt injection
Known security advisories
It lists no public code, so there is nothing to look up in GitHub Security Advisories or OSV.dev. QAmcp tests how a server behaves. It does not scan source code.
What it can do
13 tools, in the server's own words.
morningread-onlyThe day's edition: the briefing's lines by section, each with its countries, who it is about, its score (the story's main at the edition's home, 0-1) and `main…
topread-onlyThe stories of the last 24 hours in rank order, run again hourly (`as_of`, `window_hours`): each story's rank, our title, its score (0-1: the edition's perspec…
whyread-onlyWhy a piece's story is where it is on the list: its rank and score, the mains at the edition's home and at the world, the pulls that lifted it (each stake the …
searchThe stories of the last `days` whose pieces name the query, one row a story, in our words with its lead piece's outlet and link, our story page, how many piece…
save_interestMove one of the reader's magnets, what draws their own edition: a `target` and a step (away, low, normal, high) saves at once.
my_interestsread-onlyEvery magnet that draws the reader's edition, theirs over the edition's, with the words they gave, and their credits.
set_briefread-onlyThe reader's brief compiled from the record you wrote: who they are, their morning, and lanes, each standing on a target (world, country:XX, medtop:N, wd:Qn), …
answer_briefread-onlyThe question set_brief asked, answered: the choice put in the record, which is compiled again.
rehearse_briefread-onlyThe mornings of the last days the brief would have given, lane by lane: each line a happening with our title and page and the main it was ranked by; a quiet la…
save_briefThe brief kept, on the reader's yes: their active record, its lanes drawing their own edition from the next morning.
my_briefread-onlyThe brief the reader holds, with its plain-language read-back and the `confirm` an amendment saves on; or what stands in its place.
amend_briefread-onlyThe whole brief as amended by one sentence of the reader's, compiled as set_brief does, with `changes` against the one they hold.
forget_briefThe reader's brief dropped, only when they asked for that in words: nothing draws their morning but the world's.
Test results
Safe mode: only tools labeled read-only were run. How the test works
- !Every tool is describednote
9 warnings
- !morning: Inputs with no description: edition, date, audio
- !top: Inputs with no description: edition, limit
- !why: Inputs with no description: item_id, edition, date
- !save_interest: Inputs with no description: amplitude, draft, note, language
- !my_interests: Inputs with no description: edition
- !set_brief: Inputs with no description: language
- !answer_brief: Inputs with no description: language
- !rehearse_brief: Inputs with no description: language
- and 1 more
- !Follows MCP best practices (not scored yet)note
1 tip (not scored)
- !Very common tool name: search. Next to other servers' tools the AI may mix them up; a short prefix helps (like "github_search" instead of "search").
- ✓ConnectsConnected in 243 mspass
- ✓Ready for the newest MCP versionUses the latest MCP version (2026-07-28)pass
- ✓Lists its toolsFound 13 toolspass
- ✓Tools are labeled safe or dangerousEvery tool has a safety labelpass
- ✓No known hidden instructionsNone of the hidden-instruction wordings we know were found in tool descriptions. New wordings can get past this check.pass
- ✓Tools work with good inputspass
All good
- ✓morning: Worked (488 ms)
- ✓top: Worked (543 ms)
- ✓why: Answered with an error, probably because of our sample input: This story is not on the World edition list of 8 Nov.
- –my_interests: Needs you to sign in, so it was not tested (not counted against the server)
- –answer_brief: Needs you to sign in, so it was not tested (not counted against the server)
- –rehearse_brief: Needs you to sign in, so it was not tested (not counted against the server)
- –my_brief: Needs you to sign in, so it was not tested (not counted against the server)
- –amend_brief: Needs you to sign in, so it was not tested (not counted against the server)
- and 5 more
- ✓Bad inputs give a clean errorpass
All good
- ✓morning: Refused wrong type for "audio" with a clear error
- ✓top: Refused wrong type for "edition" with a clear error
- ✓why: Refused missing required input "item_id" with a clear error
- ✓why: Refused wrong type for "item_id" with a clear error
- –my_interests: Needs you to sign in, so bad inputs were not tested
- –answer_brief: Needs you to sign in, so bad inputs were not tested
- –rehearse_brief: Needs you to sign in, so bad inputs were not tested
- –amend_brief: Needs you to sign in, so bad inputs were not tested
- ✓Tools work togetherpass
2 chains working
- ✓morning → why: why found what morning returned (item_id = "550927")
- ✓top → why: why found what top returned (item_id = "572254")
- ✓Fast enoughpass
All good
- ✓Connecting took 243 ms
- ✓Listing tools took 156 ms
- ✓Average tool answer: 470 ms
- ✓top: Slowest tool: top (543 ms)
- ✓No secrets leakedChecked 19 error messages: no secrets or internal detailspass
- ✓Resources and prompts workpass
All good (1 resource, 1 prompt)
- ✓Opened 1 of 1 resources
- ✓An address that does not exist gets a clean "not found" error
- ✓Opened 1 of 1 prompts
Test it yourself
Free and open source. Same checks, same score.
npx qamcp-cli https://stablenews.ai/mcp/newsShow this score on GitHub
Free for every server in the directory. Paste this into your README: the badge shows the score from our latest test.
[](https://qamcp.dev/s/ai-stablenews-news.html)Score over time
- 8 Nov 202694
One test so far. New tests will show here.
Reviews
Coming soon: reviews from people who use Stable News, signed in with GitHub, one per person. Reviews never change the score.
QAmcp Monitor tests it every hour and emails you when it goes down, the score drops, a tool changes or a secret leaks. Free for 1 server, no card.
Have a concern, or want it removed? Ask us and we'll take it off the list. How the directory works