mirror of
https://github.com/supabase/supabase.git
synced 2026-10-10 03:45:06 +03:00
## Motivation When Assistant runs a potentially destructive tool like `execute_sql`, it stops the LLM request and prompts for client-side approval and execution of the tool. After approval, a second request kicks off under a separate trace. This has made scoring and [Topics](https://www.braintrust.dev/blog/topics) classification challenging, as the generated `output` is split across stateless requests. The [span-level scoring](https://www.braintrust.dev/docs/evaluate/custom-code#score-spans) approach we've used thusfar (after the LLM call, we massage the result into an `output` payload that's stuck onto the root span) has been cumbersome and led to invalid scores / topics where only part of the assistant response is considered. It's also inefficient, as we're duplicating potentially large info (like the `search_docs` output) that already exists within the trace. An alternative to scoring spans is to [score traces](https://www.braintrust.dev/docs/evaluate/custom-code#score-traces). Braintrust [best practices](https://www.braintrust.dev/docs/evaluate/score-online#best-practices) advise: > Use span scope for evaluating individual operations or outputs. Use trace scope for evaluating multi-turn conversations, overall workflow completion, or when your scorer needs access to the full execution context. We've also received [direct guidance](https://supabase.slack.com/archives/C05QYJBLX89/p1777925770927149?thread_ts=1777905716.911979&cid=C05QYJBLX89) from their team to use this approach. ## Changes Migrates eval scorers from custom `AssistantEvalOutput` shape to trace-level scoring via `trace.getThread()` / `trace.getSpans()`, with thread parsing that scores the full latest Assistant turn and passes prior conversation separately where relevant. Moves `execute_sql` and `deploy_edge_function` from client-side execution after approval to AI SDK `needsApproval` + server-side `execute()`. SQL results returned to the model are gated by AI opt-in level, so row data is only included with `schema_and_log_and_data`; otherwise the tool returns the no-data-permissions sentinel. Adds `metadata.isFinalStep` to disambiguate multiple LLM requests within an "assistant" turn due to tool call requests/responses. For online evals, this means we should configure automations to only score traces with `metadata.isFinalStep = true` to ensure we're judging the complete generated response. Other minor kaizen changes: - Renamed `promptProviderOptions` to `systemProviderOptions` to clarify that this is associated with the "system" message and disambiguate from the root `providerOptions` - Adds `evals/trace-utils.ts` to handle Zod validation of the `unknown` span shapes from Braintrust, to more easily access typed inputs/output on tool spans. - Bumps AI SDK floor version `^6.0.116` → `^6.0.174` - Tweaked the "Conciseness" scorer to not unfairly dock points for the new `[called tool_name]` labels in serialized assistant response ## Verification In the studio staging build, I asked Assistant to create a todos table with 3 sample todos. I manually approved the `execute_sql` call and saw Assistant generate text before & after the call. In Braintrust I verified two traces were produced (see [filtered logs](https://www.braintrust.dev/app/supabase.io/p/Assistant/logs?v=Staging&tvt=trace&search={%22filter%22:[{%22text%22:%22metadata.environment%2520%253D%2520%27staging%27%22,%22label%22:%22metadata.environment%2520%253D%2520%27staging%27%22,%22originType%22:%22btql%22},{%22text%22:%22%2560Chat%2520ID%2560%2520%253D%2520%25221cb2ac45-e5e7-458c-9da4-3bf6863b8842%2522%22,%22label%22:%22Chat%2520ID%2520equals%25201cb2ac45-e5e7-458c-9da4-3bf6863b8842%22,%22originType%22:%22form%22}]})), the first with `metadata.isFinalStep = false` and the second with `metadata.isFinalStep = true`. In the Braintrust staging scorers, I ran the preview Completeness scorer on the second trace and verified it sees the complete Assistant response including markers for tool calls ([link to trace](https://www.braintrust.dev/app/supabase.io/p/Assistant%20(Staging%20Scorers)/trace?object_type=project_logs&object_id=b5214b62-ad1e-4929-9d5b-40b1daebe948&r=0ed0a4f8-8aff-4a34-bb1d-1df1d88a5070&s=ff9015f8-6bf7-4ab3-83a9-ca4e69e27e82)) <img width="1193" height="960" alt="CleanShot 2026-05-07 at 11 27 10@2x" src="https://github.com/user-attachments/assets/509d4858-c3a1-4068-986d-3aa4d5617d1a" /> I also tested the `deploy_edge_function` workflow and verified it still prompts for permission and warns on deployment of existing functions. **References** - https://www.braintrust.dev/docs/evaluate/custom-code#score-traces - https://ai-sdk.dev/docs/ai-sdk-core/tools-and-tool-calling#tool-execution-approval Supercedes https://github.com/supabase/supabase/pull/45556 and https://github.com/supabase/supabase/pull/45339 Closes AI-473 <!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit * **New Features** * Tool actions (SQL execution, edge-function deploy) now require explicit user Approve/Deny before proceeding. * **Improvements** * Assistant pauses for approval responses before sending follow-ups, giving clearer control over risky actions. * Deploy/replace flows show confirmation and clearer replace warnings. * Evaluation/scoring updated to use richer trace data for more accurate assistant performance signals. <!-- end of auto-generated comment: release notes by coderabbit.ai -->
160 lines
4.0 KiB
TypeScript
160 lines
4.0 KiB
TypeScript
import { describe, expect, it } from 'vitest'
|
|
|
|
import { getThreadPartsFromThread } from './trace-utils'
|
|
|
|
// Sanitized mock of the thread shape returned by trace.getThread().
|
|
const MOCK_THREAD = [
|
|
{
|
|
role: 'system',
|
|
content: 'System instructions omitted for fixture.',
|
|
},
|
|
{
|
|
role: 'assistant',
|
|
content: "The user's current project is Acme Analytics.",
|
|
},
|
|
{
|
|
role: 'user',
|
|
content: 'What did we decide earlier?',
|
|
},
|
|
{
|
|
role: 'assistant',
|
|
content: [
|
|
{
|
|
type: 'text',
|
|
text: 'We decided to add an orders table with RLS policies before generating sample data.',
|
|
},
|
|
],
|
|
},
|
|
{
|
|
role: 'user',
|
|
content: 'Can you create that orders table now?',
|
|
},
|
|
{
|
|
role: 'assistant',
|
|
id: null,
|
|
content: [
|
|
{
|
|
type: 'tool_call',
|
|
tool_name: 'rename_chat',
|
|
tool_call_id: 'call_dummy_rename',
|
|
arguments: {
|
|
type: 'valid',
|
|
value: {
|
|
newName: 'Create Orders Table',
|
|
},
|
|
},
|
|
},
|
|
],
|
|
},
|
|
{
|
|
role: 'tool',
|
|
content: [
|
|
{
|
|
type: 'tool_result',
|
|
tool_name: 'rename_chat',
|
|
tool_call_id: 'call_dummy_rename',
|
|
output: {
|
|
status: 'Chat request sent to client',
|
|
},
|
|
},
|
|
],
|
|
},
|
|
{
|
|
role: 'assistant',
|
|
id: null,
|
|
content: [
|
|
{
|
|
type: 'tool_call',
|
|
tool_name: 'load_knowledge',
|
|
tool_call_id: 'call_dummy_knowledge',
|
|
arguments: {
|
|
type: 'valid',
|
|
value: {
|
|
name: 'database',
|
|
},
|
|
},
|
|
},
|
|
{
|
|
type: 'tool_call',
|
|
tool_name: 'execute_sql',
|
|
tool_call_id: 'call_dummy_sql',
|
|
arguments: {
|
|
type: 'valid',
|
|
value: {
|
|
sql: 'create table public.orders (id bigint generated by default as identity primary key);',
|
|
},
|
|
},
|
|
},
|
|
],
|
|
},
|
|
{
|
|
role: 'tool',
|
|
content: [
|
|
{
|
|
type: 'tool_result',
|
|
tool_name: 'load_knowledge',
|
|
tool_call_id: 'call_dummy_knowledge',
|
|
output: 'Knowledge fixture omitted.',
|
|
},
|
|
{
|
|
type: 'tool_result',
|
|
tool_name: 'execute_sql',
|
|
tool_call_id: 'call_dummy_sql',
|
|
output: {
|
|
type: 'text',
|
|
text: 'SQL executed successfully.',
|
|
},
|
|
},
|
|
],
|
|
},
|
|
{
|
|
role: 'assistant',
|
|
id: null,
|
|
content:
|
|
'I created the public.orders table. You should add RLS policies before exposing it to users.',
|
|
},
|
|
]
|
|
|
|
describe('getThreadPartsFromThread', () => {
|
|
it('parses a sanitized Braintrust trace.getThread payload', () => {
|
|
expect(getThreadPartsFromThread(MOCK_THREAD)).toEqual({
|
|
projectContext: "The user's current project is Acme Analytics.",
|
|
priorConversation:
|
|
'[user]\nWhat did we decide earlier?\n\n[assistant]\nWe decided to add an orders table with RLS policies before generating sample data.',
|
|
currentUserInput: 'Can you create that orders table now?',
|
|
lastAssistantTurn:
|
|
'[assistant]\n[called rename_chat]\n\n[assistant]\n[called load_knowledge]\n[called execute_sql]\n\n[assistant]\nI created the public.orders table. You should add RLS policies before exposing it to users.',
|
|
})
|
|
})
|
|
|
|
it('uses the most recent project context message', () => {
|
|
expect(
|
|
getThreadPartsFromThread([
|
|
{
|
|
role: 'assistant',
|
|
content: "The user's current project is Old Project.",
|
|
},
|
|
...MOCK_THREAD,
|
|
])
|
|
).toMatchObject({
|
|
projectContext: "The user's current project is Acme Analytics.",
|
|
})
|
|
})
|
|
|
|
it('returns prior conversation without current turn parts when there is no user message', () => {
|
|
expect(
|
|
getThreadPartsFromThread([
|
|
{
|
|
role: 'assistant',
|
|
content: 'I can help with your Supabase project.',
|
|
},
|
|
])
|
|
).toEqual({
|
|
projectContext: null,
|
|
priorConversation: '[assistant]\nI can help with your Supabase project.',
|
|
currentUserInput: null,
|
|
lastAssistantTurn: null,
|
|
})
|
|
})
|
|
})
|