Skip to content

Context7 Platform -- Up-to-date docs for AI agents

CTX7-3282: Fix data privacy docs for Enterprise query storage and LLM providers (#3484)

Enes Gulesa989fd49e69c
diff --git a/docs/enterprise.mdx b/docs/enterprise.mdx
index 5ebde6f3..7cb13bbd 100644
--- a/docs/enterprise.mdx
+++ b/docs/enterprise.mdx
@@ -59,7 +59,7 @@ Context7 runs on **SOC 2 Type II** compliant infrastructure provided by [Upstash
- **ISO 27001** certification in progress
- Open-source MCP server — publicly auditable at [github.com/upstash/context7](https://github.com/upstash/context7)
-Enterprise customers can also disable query storage, use their own LLM provider for code extraction and private library ranking, and limit context retrieval to privately indexed documentation only.
+Context7 does not store queries from Enterprise teamspaces for benchmarking. Enterprise customers can also use their own LLM provider for code extraction and private library ranking, and limit context retrieval to privately indexed documentation only.
For full details, see the [Security](/security/overview) page and the [Upstash Trust Center](https://trust.upstash.com/).
diff --git a/docs/security/data-privacy.mdx b/docs/security/data-privacy.mdx
index 84a1dfdc..06c73a24 100644
--- a/docs/security/data-privacy.mdx
+++ b/docs/security/data-privacy.mdx
@@ -37,17 +37,17 @@ The search queries formulated by the MCP client (not your original prompts) are
**Documentation Reranking**
-MCP-formulated queries are passed to LLMs to rerank and select the most relevant documentation for your request. Context7 uses well-known, trusted LLM providers for this purpose — including **OpenAI**, **Google Gemini**, and **Anthropic**.
+Context7 sends MCP-formulated queries to LLMs, which rerank the documentation and pick the most relevant results for your request. Context7 uses only trusted LLM providers that do not train on the data it sends them.
**Benchmarking and Quality Improvement**
-MCP-formulated queries are anonymously stored and used to benchmark retrieval accuracy and improve the documentation matching pipeline over time.
+Context7 stores MCP-formulated queries anonymously and uses them to measure and improve retrieval accuracy. Context7 does not store queries from Enterprise teamspaces for benchmarking.
### Enterprise Controls
- On-premise Enterprise plans can use their own LLM provider for code extraction and private library ranking
- On-premise Enterprise plans can disable public documentation usage, limiting context retrieval to privately indexed documentation only
-- Enterprise plans can disable query storage for benchmarking — however, this may affect the quality of context retrieval over time
+- Queries from Enterprise teamspaces are never stored for benchmarking. No setting is needed
Contact our sales team at [context7.com/contact](https://context7.com/contact) for Enterprise and on-premise plan details.
@@ -108,7 +108,7 @@ Restrict which libraries are accessible to your teamspace. You can choose betwee
- Library documentation
- Code examples from documentation
- Metadata about indexed libraries
-- Queries formulated by the MCP client
+- Queries formulated by the MCP client, used for benchmarking (not stored for Enterprise teamspaces)
**What we don't store:**