feat: add model_intent option to MCP server configs (#23717)

Add a per-MCP-server `model_intent` toggle that wraps tool schemas with
a
`model_intent` field, requiring the LLM to provide a human-readable
description of each tool call's purpose. The intent string is shown as a
status label in the UI instead of opaque tool names, and is
transparently
stripped before the call reaches the remote MCP server.

Built-in tools have rich specialized renderers (terminal blocks, file
diffs,
etc.) and don't need this. MCP tools hit `GenericToolRenderer` which
only
shows raw tool names and JSON — that's where model_intent adds value.

The model learns what to provide via the JSON Schema `description` on
the
`model_intent` property itself — no system prompt changes needed.

<details>
<summary>Implementation details</summary>

### Architecture

Inspired by the `withModelIntent()` pattern from `coder/blink`, adapted
for
Go + React. The wrapping is entirely in the `mcpclient` layer — tool
implementations never see `model_intent`.

**Schema wrapping** (`mcpToolWrapper.Info()`): When enabled, wraps the
original tool parameters under a `properties` key and adds a
`model_intent`
string field with a rich description that teaches the model inline.

**Input unwrapping** (`mcpToolWrapper.Run()`): Strips `model_intent` and
unwraps `properties` before forwarding to the remote MCP server. Handles
three input shapes models may produce:
1. `{ model_intent, properties: {...} }` — correct format
2. `{ model_intent, key: val, ... }` — flat, no wrapper
3. Malformed — falls through gracefully

**Frontend extraction**: `streamState.ts` extracts `model_intent` from
incrementally parsed streaming JSON. `messageParsing.ts` extracts it
from
persisted tool call args.

**UI rendering**: `GenericToolRenderer` shows the capitalized intent
string
as the primary label when available, falling back to the raw tool name.

### Changes
- Database: `model_intent` boolean column on `mcp_server_configs`
- SDK: `ModelIntent` field on config/create/update types
- API: pass-through in create/update handlers + converter
- mcpclient: schema wrapping in `Info()`, input unwrapping in `Run()`
- Frontend: extraction from streaming + persisted args
- UI: intent label in `GenericToolRenderer`, toggle in admin panel
- Tests: 6 new tests (schema wrapping, unwrapping, passthrough,
fallback)

### Decision log
- **Option lives on MCPServerConfig, not model config**: Built-in tools
  already have rich renderers; only MCP tools benefit from model_intent.
- **No system prompt changes**: The JSON Schema `description` on the
  `model_intent` property teaches the model inline.
- **Pointer bool on update request**: Follows existing pattern (`*bool`)
  so PATCH requests don't reset the value when omitted.

</details>
This commit is contained in:
Kyle Carberry
2026-03-27 14:23:25 +00:00
committed by GitHub
parent 50c0c89503
commit d973a709df
21 changed files with 402 additions and 32 deletions
+6 -3
View File
@@ -64,9 +64,10 @@ type MCPServerConfig struct {
// Availability policy set by admin.
Availability string `json:"availability"` // "force_on", "default_on", "default_off"
Enabled bool `json:"enabled"`
CreatedAt time.Time `json:"created_at" format:"date-time"`
UpdatedAt time.Time `json:"updated_at" format:"date-time"`
Enabled bool `json:"enabled"`
ModelIntent bool `json:"model_intent"`
CreatedAt time.Time `json:"created_at" format:"date-time"`
UpdatedAt time.Time `json:"updated_at" format:"date-time"`
// Per-user state (populated for non-admin requests).
AuthConnected bool `json:"auth_connected"`
@@ -97,6 +98,7 @@ type CreateMCPServerConfigRequest struct {
Availability string `json:"availability" validate:"required,oneof=force_on default_on default_off"`
Enabled bool `json:"enabled"`
ModelIntent bool `json:"model_intent"`
}
// UpdateMCPServerConfigRequest is the request to update an MCP server config.
@@ -124,6 +126,7 @@ type UpdateMCPServerConfigRequest struct {
Availability *string `json:"availability,omitempty" validate:"omitempty,oneof=force_on default_on default_off"`
Enabled *bool `json:"enabled,omitempty"`
ModelIntent *bool `json:"model_intent,omitempty"`
}
func (c *Client) MCPServerConfigs(ctx context.Context) ([]MCPServerConfig, error) {