Remove system role references from docs (#3356)

The MCP spec only supports "user" and "assistant" roles for prompt
messages. Fixes #3160.

Co-authored-by: Marvin Context Protocol <41898282+Marvin Context Protocol@users.noreply.github.com>
Co-authored-by: Jeremiah Lowin <jlowin@users.noreply.github.com>
This commit is contained in:
Jeremiah Lowin 2026-03-02 16:36:24 -05:00 committed by GitHub
commit 2ab8beb195
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
3 changed files with 7 additions and 7 deletions

View file

@ -90,9 +90,9 @@ async with client:
"expertise": "python programming"
})
# Typically returns messages with role="system"
system_message = result.messages[0]
print(f"System prompt: {system_message.content}")
# Access the returned messages
message = result.messages[0]
print(f"Prompt: {message.content}")
```
Conversation templates generate multi-turn flows:

View file

@ -116,7 +116,7 @@ result = await client.call_tool(
## Message Format
Rendered prompts return a messages array following the standard MCP format. Each message includes:
- `role`: The message role (typically "user", "assistant", or "system")
- `role`: The message role ("user" or "assistant")
- `content`: The message text content
Multi-message prompts are supported - the array will contain all messages in order.

View file

@ -190,9 +190,9 @@ async with client:
"expertise": "python programming"
})
# Typically returns messages with role="system"
system_message = result.messages[0]
print(f"System prompt: {system_message.content}")
# Access the returned messages
message = result.messages[0]
print(f"Prompt: {message.content}")
```
### Conversation Templates