Hello, AI
com.helloai/directoryhelloai.comstreamable-httpv1.0.0Scanned
Ask your assistant which frontier model to use and get a live, Elo-backed answer.
6 tools: 1 change things and 5 only read.
- Reads private data · 0
- Reads text others write · 1
- Can send data out · 0
- No findings
Anlyon alternative
1 of these tools change something. Run those as governed actions.
Declare what an action changes, and Anlyon puts it under a shared limit, binds every approval to the exact request, and grades the outcome. It works on any HTTPS API you can describe as an action.
Anlyon does not run, proxy or govern this MCP server. A tool your own code calls directly is not routed through Anlyon and is not governed by it.
Import an OpenAPI spec →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.
6 tools
Changes things 1Claude Code: ask
recommend_modelsRank Hello, AI's tracked frontier models for a task.No findings
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.
Description, as the server wrote it Rank Hello, AI's tracked frontier models for a task. Same scoring as GET /api/recommend: task match 40%, Elo 35%, input cost 15%, context 10% (Elo 55%, cost 25%, context 20% when no task matches). Only rated models rank; unrated ones are listed separately. score is an ordering aid, not a quality measure.
Annotations the server declares
- readOnlyHint: true
- idempotentHint: true
- openWorldHint: false
Only reads 5Show 5 read-only tools
get_modelOne tracked model's full record by id (claude, fable, muse, gemini, qwen, grok) or exact name.No findings
Why it is in this group
- The name "get_model" says "get": a word for reading.
Description, as the server wrote it One tracked model's full record by id (claude, fable, muse, gemini, qwen, grok) or exact name.
Annotations the server declares
- readOnlyHint: true
- idempotentHint: true
- openWorldHint: false
get_statusData freshness, app version, model and article counts, the scoring version and snapshot fingerprint, and the rate limit.No findings
Why it is in this group
- The name "get_status" says "get": a word for reading.
Description, as the server wrote it Data freshness, app version, model and article counts, the scoring version and snapshot fingerprint, and the rate limit.
Annotations the server declares
- readOnlyHint: true
- idempotentHint: true
- openWorldHint: false
list_articlesRecent Hello, AI articles, newest first: title, date, category, excerpt and URL.Reads text others writeNo findings
Why it is in this group
- The name "list_articles" says "list": a word for reading.
What it can reach
- untrusted-inputThe description's first sentence says "url": it reads web pages, which anyone can write.
Description, as the server wrote it Recent Hello, AI articles, newest first: title, date, category, excerpt and URL.
Annotations the server declares
- readOnlyHint: true
- idempotentHint: true
- openWorldHint: false
list_categoriesThe curated use-case categories with their curator-chosen leader and a one-paragraph insight.No findings
Why it is in this group
- The name "list_categories" says "list": a word for reading.
Description, as the server wrote it The curated use-case categories with their curator-chosen leader and a one-paragraph insight. The leader is a hand-set label, not derived from Elo.
Annotations the server declares
- readOnlyHint: true
- idempotentHint: true
- openWorldHint: false
list_modelsAll tracked models with full records: Elo and its provenance (elo_source), prices per million tokens, context window, strengths, description.No findings
Why it is in this group
- The name "list_models" says "list": a word for reading.
Description, as the server wrote it All tracked models with full records: Elo and its provenance (elo_source), prices per million tokens, context window, strengths, description. rated is false when the stored Elo is a predecessor's, missing or stale.
Annotations the server declares
- readOnlyHint: true
- idempotentHint: true
- openWorldHint: false
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 write1
- 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.
Findings
Matches of fixed checks, with the text they matched.
No check matched anything in these tool definitions. That is not a safety guarantee.
History
Changes to its tool definitions and to its remote endpoint, newest first.
No change seen yet. Its tools are rescanned every day, and a remote endpoint is looked at once a week.
Install
Pinned to version 1.0.0. Values in angle brackets are yours to fill in.
claude mcp add --transport http directory https://helloai.com/mcpclaude mcp add --transport http directory https://helloai.com/mcpRun 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.
{
"permissions": {
"allow": [
"mcp__directory__list_models",
"mcp__directory__get_model",
"mcp__directory__list_categories",
"mcp__directory__list_articles",
"mcp__directory__get_status"
],
"ask": [
"mcp__directory__recommend_models"
],
"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 directory. Review it before you use it.
Capability tags behind 1 rule
- mcp__directory__list_articles 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://anlyon.dev/api/v1/servers/com.helloai/directory/verify?hash=1f98d54db59cc541cf3643829d3774621beb31de5838639898a081454c4804df" | jq -e '.changed == false'Pinned hash 1f98d54db59c, from Oct 27, 2026, 16:18 UTC.
Watch
Get told when its tools change: a new tool, a new description, a new class.
https://anlyon.dev/servers/com.helloai/directory/changes.xmlAn Atom feed. Works in any feed reader or Slack's RSS app.
Badge
[](https://anlyon.dev/servers/com.helloai/directory)Source: MCP Registry. Synced Oct 27, 2026, 16:17 UTC.