AI News
PYPI · AI-NEWS-MCP
Collects AI news from many Sources into a local store for Claude to query. No API key.
Available components
Last scanned (18 hours ago)
Scores higher than 82% of PyPI servers
Category News and Social
How this component scores in each security and reliability category. Every signal is checked automatically from public evidence about the published package, including repeated runs of it in an isolated sandbox, and we only credit what we can confirm. How we score → Why this is hard to score →
Supply Chain Security100
- No malware found by supply-chain analysis.Pass
- No known CVEs affecting this package version or its production dependencies.Pass
- Runs uv_build at install time, a recognised build step with no custom scripting around it. View diagnostics → Pass
- 2 of 48 dependencies flagged as unhealthy. View diagnostics → Partial
Provenance & Transparency100
- Source repository is publicly reachable at the declared URL. View diagnostics → Pass
- Cryptographically verified build provenance (signed, bound to duncan-wen/ai-news-mcp). View diagnostics → Pass
- Clear OSI-approved license (MIT).Pass
- Actively maintained (last published 3 days ago).Pass
- Publishes a security disclosure policy (SECURITY.md).Pass
Schema Quality & AI Usability64
- AI-judged instruction clarity (excellent).Pass
- Context-footprint check failed: tool/resource definitions use about 1185 tokens (~237/item across 5 items; 5 tools + 0 resources), over budget; trim descriptions and params. See how to fix → Fail
- Usage-examples check failed: none of the tools include examples. See how to fix → Fail
Stability & Change Management33
- Stability observed for 10 of 30 days with no destabilising changes; credit accrues until the full window elapses.Partial
Tool Coverage71
- 100% of tools have a non-trivial description (not blank, and not just the tool's name).Pass
- 0% of tool parameters carry a description.Fail
- Structured output schemas are declared (100% of tools); any adoption earns full credit.Pass
Tool Safety100
- No prompt-injection markers were found in the server instructions, tool names or descriptions we captured.Pass
- We read all 5 captured tool definition(s), and no name or description among them implies an irreversible operation.Pass
- An AI judge read all 5 captured unit(s) of tool text and found none that tries to manipulate the model reading it.Pass
Capabilities100
- Implements a current MCP spec version (2026-07-28).Pass
Checks we ran (16)
- Malware scan
- Known CVEs
- Install scripts
- Dependency health
- Source repository
- Provenance
- License
- Maintenance
- Security disclosure
- Schema quality
- Stability
- Tool coverage
- Injection markers
- Destructive annotations
- Judged manipulation
- MCP protocol
Does not apply to this package (2)
- Endpoint SecurityScored on hosted endpoints only. A package has no live endpoint of its own for us to probe.
- Transport & ReachabilityScored on hosted endpoints only. A package has no live endpoint of its own for us to probe.
The score does not measure everything. What the score does not measure →
Every change we have recorded for this component, newest first. Security-relevant changes are always shown. ▲ marks a change for the better, ▼ a change for the worse; unmarked changes are neutral.
- 21 Nov 26 +1
No change was recorded against any check on this day. Stability & Change Management went from 30 to 33. That category is still filling its 30-day observation window: 9 days of observed history at the previous scan, 10 at this one. The score rises as the window fills, whether or not the server changes.
- 19 Nov 26 +26
- Malware scan: unverified → pass ▲ security
- Provenance: unverified → pass ▲ security
- The attested source repository moved: duncan-wen/ai-news-mcp security
- 18 Nov 26 −25
- Provenance: pass → unverified ▼ security
- Malware scan: pass → unverified ▼ security
- The attested source repository moved: duncan-wen/ai-news-mcp security
- Package version: 0.2.1 → 0.3.0 functional
- 16 Nov 26 +16
- Malware scan: unverified → pass ▲ security
- 15 Nov 26 −15
- Malware scan: pass → unverified ▼ security
- Package version: 0.2.0 → 0.2.1 functional
- 14 Nov 26 +1
No change was recorded against any check on this day. Stability & Change Management went from 7 to 10. That category is still filling its 30-day observation window: 2 days of observed history at the previous scan, 3 at this one. The score rises as the window fills, whether or not the server changes.
- 13 Nov 26 +1
- Stability: unverified → 0.07 ▲ functional
- 12 Nov 26 +15
- Malware scan: unverified → pass ▲ security
Diagnostic detail from the automated scan of this channel: what the scanner observed at each step, so you can see exactly where a check passed or failed. It is informational only and never changes the trust score.
Captured 21 Nov 2026 · Analysed pypi/ai-news-mcp@0.3.0
Provenance Verified
A signed build attestation was found and verified, binding this exact artifact to the source repository it claims to come from.
| Result | Verified |
|---|---|
| Ecosystem | pypi |
| Reason | Verified |
| Discovered via | Registry attestation endpoint |
| Source repo | duncan-wen/ai-news-mcp |
| Certificate issuer | https://token.actions.githubusercontent.com |
| Certificate SAN | https://github.com/duncan-wen/ai-news-mcp/.github/workflows/publish.yml@refs/tags/v0.3.0 |
| Rekor log index | 3956854659 |
| Predicate type | PyPI publish attestation https://docs.pypi.org/attestations/publish/v1 |
| Subject digest | sha256:080b7622c06de0b74d045b01714037a275f427c98e3c329868482f41967646ff |
Background: How many MCP packages publish verified provenance →
Install scripts 1 script
| Hook | Tier | Command |
|---|---|---|
| build_backend | Known build tool | uv_build |
Background: Why install scripts are a supply-chain risk →
Dependencies 48 packages
| Packages resolved | 48 |
|---|---|
| Stale | 1 |
| No linked repository | 1 |
| Tree resolution | Complete |
Background: SBOMs and build attestations, explained →
The tools this component advertises to a client, with an estimated token cost for each. Expand a tool to see its parameters and schema. The per-tool counts are indicative and are not scored directly; the schema's total context footprint is one signal in Schema Quality & AI Usability. A tool's description is untrusted text the model reads on every call, which is what makes this list a security surface and not just an inventory: how tool poisoning works →
collect_now ~76
Collect new Items from every Source now. Use only when the user asks to refresh or collect now: `whats_new` already collects by itself when the data is over 24 hours old. Takes about a minute. Reports how many Items each Source added (0 for a failed one), and which Sources failed and why.
Input schema present but exposes no named parameters.
| Name | Type | Req | Description |
|---|---|---|---|
| failed | array | yes | – |
| sources | array | yes | – |
No examples provided.
list_sources ~84
Every Source with its URL, default Categories and collection state. The state is when it was last tried, when it last succeeded, its last error and how many collections in a row it has failed. The last error is kept after a success, so with 0 failures in a row it is from before `last_success_at`. Use it to check which Sources work.
Input schema present but exposes no named parameters.
| Name | Type | Req | Description |
|---|---|---|---|
| sources | array | yes | – |
No examples provided.
read_item ~156
The full text of an Item: fetches its original page and extracts the main text. Use when an Item's summary is not enough. `id` is an Item's `id` from `whats_new` or `search`. `text` is the first `max_length` characters (default 10,000), with no marker added; `truncated` says whether that left some out and `full_length` gives the whole text's length, so call again with a larger `max_length` for the rest. If the page cannot be fetched or has no main text, the error gives the Item's summary and link instead.
| Name | Type | Req | Description |
|---|---|---|---|
| id | integer | yes | – |
| max_length | integer | – | – |
| Name | Type | Req | Description |
|---|---|---|---|
| full_length | integer | yes | – |
| id | integer | yes | – |
| text | string | yes | – |
| title | string | yes | – |
| truncated | boolean | yes | – |
| url | string | yes | – |
No examples provided.
search ~357
Search stored Items by text, publish time, Category and Source, in any mix. Use for a question about a subject or time span, such as last week's Engineering Items about MCP or what was published today, and whenever `whats_new` is not clearly asked for. Every filter is optional; an omitted one matches every Item. `query`: the title or the whole summary contains it (case-insensitive). `since` / `until`: published at or after `since` and before `until`, ISO 8601; a time without an offset, or a date alone (midnight), is UTC. `categories`: in any of them (News, Applications, Engineering, Models; see `whats_new`). `sources`: from any of these Source ids (see `list_sources`). An empty `categories` or `sources` list matches nothing. Results are newest first, up to `limit` (default 20); Items published at the same time are never split, so a few more or fewer can be shown. `more` counts the matching Items left out (0: none); for the next page, call again with the same filters and `until` the oldest shown Item's `published_at`. A summary shown is plain text cut to `summary_length`, about 500 characters by default; a shorter length skims many Items by title, 0 shows none, and `read_item` gives the full text. Searching never moves a Checkpoint.
| Name | Type | Req | Description |
|---|---|---|---|
| categories | – | – | – |
| limit | integer | – | – |
| query | – | – | – |
| since | – | – | – |
| sources | – | – | – |
| summary_length | integer | – | – |
| until | – | – | – |
| Name | Type | Req | Description |
|---|---|---|---|
| items | array | yes | – |
| more | integer | – | – |
No examples provided.
whats_new ~512
What's New: Items collected since the last Checkpoint and published in the last 7 days. Use only when the user asks for what is new since they last looked: what's new in AI or in some Categories, anything new, today's briefing. Calling it marks Items as seen and cannot be undone, so for anything else use `search`, which never does: a time span ("today", "recently", "this week"), a subject ("what's new with MCP"), or when unsure. Each Category has its own Checkpoint; only the Checkpoints of the Categories asked about (omitted: all) move to now, so Items shown once are not shown again. Every client of this server shares the Checkpoints. `previous_checkpoints` gives each asked Category's Checkpoint before this call (null: never viewed). If `items` is empty, tell the user when they last looked and offer `search` for recently published Items. Results are newest first, up to `limit` (default 60); Items published at the same time are never split, so a few more or fewer can be shown. `more` counts the Items left out; they count as seen too, so tell the user and offer `search` with the same `categories`, `since` 7 days ago and `until` the oldest shown Item's `published_at`. A summary shown is plain text cut to `summary_length`, about 500 characters by default; a shorter length skims many Items by title, 0 shows none, and `read_item` gives the full text. Categories (an Item can be in several): News: Events in the AI industry, such as launches, funding, policy, and people changes. Applications: New AI products or tools that you can use right away. Engineering: Content about how to build AI systems, such as technical posts, frameworks, lessons from practice, and releases. Models: Newly released or newly listed AI models, and changes to what they can do and what they cost. Academic papers and theory are not included. If the data is over 24 hours old this collects first, which can take about a minute; `note` says when the collection was still running. `warning` names Sources…
| Name | Type | Req | Description |
|---|---|---|---|
| categories | – | – | – |
| limit | integer | – | – |
| summary_length | integer | – | – |
| Name | Type | Req | Description |
|---|---|---|---|
| items | array | yes | – |
| more | integer | – | – |
| note | – | – | – |
| previous_checkpoints | object | yes | – |
| warning | – | – | – |
No examples provided.
How do I install the AI News MCP server?
AI News runs locally as a PyPI package, launched with uvx ai-news-mcp. Ready-made configuration for Claude, Cursor, VS Code, Codex and 5 more is on this page, copied from each client's own documentation.
pypi · ai-news-mcp
claude mcp add duncan-wen-ai-news-mcp -- uvx ai-news-mcp
What is the AI News MCP server?
AI News is an MCP server listed in the public MCP registry as io.github.duncan-wen/ai-news-mcp. Collects AI news from many Sources into a local store for Claude to query. No API key. This page covers its PyPI package (ai-news-mcp).
Is the AI News MCP server safe to use?
AI News scores 82 out of 100 on VerifyMCP. We found no known CVEs affecting it as of 21 November 2026. Its build provenance is signed and verified. That is a record of what we were able to check automatically, not an endorsement. The category breakdown on this page shows every signal behind the number, including the ones we could not confirm.
What tools does the AI News MCP server expose?
AI News exposes 5 tools: collect_now, list_sources, search, read_item, whats_new. Their descriptions and schemas cost roughly 1,185 tokens of context every time the server is loaded.
Is the AI News MCP server still maintained?
AI News is still listed as active in the MCP registry. We last reached this channel on 21 November 2026. Those dates come from our own scans of the registry and the channel itself, not from anything the publisher announced.
What licence is the AI News MCP server under?
AI News declares the MIT licence, which is OSI-approved. That covers the source only, and says nothing about the cost of any service it calls.
What did VerifyMCP test on the AI News MCP server?
VerifyMCP ran 16 checks on AI News, covering Malware scan, Known CVEs, Install scripts, Dependency health and 12 more. Endpoint Security and Transport & Reachability do not apply to a PyPI package, so they were not tested. The score does not measure everything, and the scoring documentation lists what it leaves out.