🎩 You're Invited:Meet the Socket team at Black Hat in Las Vegas, August 3-6.RSVP →
@llmintel/mcp
Advanced tools
@llmintel/mcp - npm Package Compare versions
+2
-1
| { | ||
| "name": "@llmintel/mcp", | ||
| "version": "0.1.0", | ||
| "version": "0.1.1", | ||
| "mcpName": "ai.llmintel/llmintel", | ||
| "description": "MCP server for AI model lifecycle data — check whether a model id is deprecated or retiring, and what to migrate to, across OpenAI, Anthropic, Azure, AWS Bedrock, Google, and Cohere.", | ||
@@ -5,0 +6,0 @@ "keywords": [ |
+32
-28
@@ -7,15 +7,17 @@ # @llmintel/mcp | ||
| LLMs are trained on a snapshot of the world and will confidently write `gpt-4-32k` into your code | ||
| long after it stops answering. This server gives the agent a live lookup: **is this model | ||
| deprecated, when does it stop working, and what replaces it**. Answers are normalized across | ||
| long after it stops answering. This server gives the agent a live lookup for whether a model is | ||
| deprecated and when it stops working. It returns the replacement too. Answers are normalized across | ||
| OpenAI, Anthropic, Azure AI Foundry, AWS Bedrock, Google, and Cohere, and parsed from each | ||
| provider's own deprecation pages. | ||
| **No API key. No signup.** The catalog is public. | ||
| No API key, no signup. The catalog is public. | ||
| ## Install | ||
| Add it to any MCP host. The package runs straight from npm via `npx`: | ||
| Add it to any MCP host. The package runs straight from npm via `npx`. | ||
| **Cursor**, in `.cursor/mcp.json`: | ||
| ### Cursor | ||
| In `.cursor/mcp.json`: | ||
| ```json | ||
@@ -32,3 +34,3 @@ { | ||
| **Claude Code**: | ||
| ### Claude Code | ||
@@ -39,13 +41,15 @@ ```bash | ||
| **Claude Desktop**: the same shape as the Cursor block above, in `claude_desktop_config.json`. | ||
| ### Claude Desktop | ||
| Same shape as the Cursor block above, in `claude_desktop_config.json`. | ||
| ## Tools | ||
| | Tool | Answers | | ||
| | Tool | Returns | | ||
| | --- | --- | | ||
| | `check_model` | "Is `gpt-4o` safe to use?" Returns lifecycle state, the retirement deadline in days, the replacement, and the source link. | | ||
| | `list_retiring_models` | "What breaks in the next 90 days?" Past-due models are listed first, then upcoming ones soonest-first. | | ||
| | `suggest_replacement` | "What do I move to?" The provider's own recommendation, or same-provider active models when none was published. | | ||
| | `search_models` | "What can I use from Anthropic right now?" Filters by provider and lifecycle state. | | ||
| | `recent_lifecycle_changes` | "What was deprecated this month?" The change feed across all providers. | | ||
| | `check_model` | Whether one model id is safe to use: lifecycle state, the retirement deadline in days, the replacement, and the source link. | | ||
| | `list_retiring_models` | What breaks in the next 90 days. Past-due models are listed first, then upcoming ones soonest-first. | | ||
| | `suggest_replacement` | The provider's own recommendation for what to move to. Falls back to same-provider active models when none was published. | | ||
| | `search_models` | Catalog search filtered by provider and lifecycle state. | | ||
| | `recent_lifecycle_changes` | The change feed across all providers, for questions like "what was deprecated this month". | | ||
@@ -75,21 +79,21 @@ ### Example | ||
| Deadlines are always given in **days**, because a model cannot reliably judge whether `2026-07-30` | ||
| is soon. | ||
| Deadlines are always given in days, because a model cannot reliably judge whether `2026-07-30` is | ||
| soon. | ||
| ## Design notes | ||
| **A failed lookup is never a safety verdict.** If the catalog is unreachable, the tool returns an | ||
| MCP error and says so. An agent that read a network failure as "no deprecation found" would happily | ||
| ship a retired model id. A model that simply isn't tracked gets the same treatment: it returns "not | ||
| in the catalog, verify with the provider", never "OK". | ||
| A failed lookup is never a safety verdict. If the catalog is unreachable, the tool returns an MCP | ||
| error and says so. An agent that read a network failure as "no deprecation found" would happily ship | ||
| a retired model id. A model that simply isn't tracked gets the same treatment: it returns "not in the | ||
| catalog, verify with the provider", never "OK". | ||
| **Aliases resolve.** Pass whatever string is literally in the code (`gpt-4o`, | ||
| `anthropic/claude-opus-4-1`, `azure/gpt-4o`) and it resolves to the canonical tracked model. | ||
| Pass whatever string is literally in the code (`gpt-4o`, `anthropic/claude-opus-4-1`, `azure/gpt-4o`) | ||
| and it resolves to the canonical tracked model. | ||
| **Provider-stated recommendations are labelled separately from ours.** When the provider's own | ||
| deprecation notice names a successor, that is what you get. When it doesn't, the fallback list of | ||
| same-provider active models is marked as candidates to evaluate. | ||
| When the provider's own deprecation notice names a successor, that is what you get. Otherwise the | ||
| fallback list of same-provider active models is labelled as candidates to evaluate, so an agent can | ||
| tell the two apart. | ||
| **Past-due models are separated out.** Anything past its retirement date is broken now, so it gets | ||
| its own heading instead of sitting in "retiring soon". | ||
| Anything past its retirement date is broken now, so it gets its own heading instead of sitting in | ||
| "retiring soon". | ||
@@ -105,3 +109,3 @@ ## Configuration | ||
| Anonymous callers get 30 requests/minute per IP, which is ample for interactive agent use. | ||
| Anonymous callers get 30 requests/minute per IP, enough for interactive agent use. | ||
@@ -112,3 +116,3 @@ ## Data provenance | ||
| lifecycle term (`sourceTerm`), so a normalization decision is always auditable. Changes go through a | ||
| human verification queue before publication, and collector freshness is public at | ||
| human verification queue before publication. Collector freshness is public at | ||
| [`/v1/status`](https://llmintel.ai/v1/status). | ||
@@ -115,0 +119,0 @@ |
+4
-4
| { | ||
| "$schema": "https://static.modelcontextprotocol.io/schemas/2025-09-29/server.schema.json", | ||
| "$schema": "https://static.modelcontextprotocol.io/schemas/2025-12-11/server.schema.json", | ||
| "name": "ai.llmintel/llmintel", | ||
| "description": "Check whether an LLM model id is deprecated, retiring, or retired — and what to migrate to. Normalized lifecycle data across OpenAI, Anthropic, Azure, AWS Bedrock, Google, and Cohere, sourced from each provider's own deprecation pages.", | ||
| "version": "0.1.0", | ||
| "description": "Check whether an LLM model id is deprecated, retiring, or retired, and what to migrate to.", | ||
| "version": "0.1.1", | ||
| "websiteUrl": "https://llmintel.ai", | ||
@@ -16,3 +16,3 @@ "repository": { | ||
| "identifier": "@llmintel/mcp", | ||
| "version": "0.1.0", | ||
| "version": "0.1.1", | ||
| "transport": { "type": "stdio" }, | ||
@@ -19,0 +19,0 @@ "environmentVariables": [ |
No alert changes
Improved metrics
- Number of lines in readme file
131
3.15%Worsened metrics
- Total package byte prevSize
130266
-0.17%No dependency changes