@wix/evalforge-evaluator 0.178.0 → 0.179.0

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (12) hide show
  1. package/README.md +13 -2
  2. package/build/index.js +6911 -239
  3. package/build/index.js.map +4 -4
  4. package/build/index.mjs +6940 -246
  5. package/build/index.mjs.map +4 -4
  6. package/build/types/ambassador-converters.d.ts +68 -0
  7. package/build/types/api-client.d.ts +26 -3
  8. package/build/types/config.d.ts +12 -1
  9. package/build/types/run-scenario/agents/shared/trace-emit.d.ts +5 -2
  10. package/build/types/run-scenario/index.d.ts +4 -2
  11. package/build/types/run-scenario/run-agent-with-context.d.ts +4 -2
  12. package/package.json +13 -5
package/README.md CHANGED
@@ -31,14 +31,25 @@ evaluator <project-id> <eval-run-id>
31
31
  | `AI_GATEWAY_HEADERS` | No | Custom headers for AI Gateway (newline-separated `key:value` pairs) |
32
32
  | `EVAL_API_PREFIX` | No | API path prefix (e.g., `/api/v1`) |
33
33
  | `EVALUATIONS_DIR` | No | Directory for evaluation working directories |
34
- | `TRACE_PUSH_URL` | No | URL for pushing trace events (remote job execution) |
34
+ | `TRACE_PUSH_URL` | No | Enables remote trace push when set (remote job execution). Events are pushed via the gRPC `PushTraceEvent` RPC; the URL value itself is legacy |
35
35
  | `EVAL_ROUTE_HEADER` | No | `x-wix-route` header for deploy preview routing |
36
- | `EVAL_AUTH_TOKEN` | No | Bearer token for public endpoint authentication |
36
+ | `EVAL_AUTH_TOKEN` | No | Bearer token for the remaining legacy REST public endpoints |
37
+ | `EVAL_GRPC_AUTH_TOKEN` | No | S2S-signed token for ambassador/gRPC calls (absent in local dev — calls go out unauthenticated) |
37
38
 
38
39
  For OpenCode runs, the evaluator sets `lsp: false` in `OPENCODE_CONFIG_CONTENT` and `OPENCODE_DISABLE_LSP_DOWNLOAD` / `OPENCODE_DISABLE_FILETIME_CHECK` in the process environment (same as ditto `codegen`) to avoid LSP hangs after edit tools and spurious "file modified since last read" failures in automated evals.
39
40
 
40
41
  The evaluator is typically launched by the backend (locally or on a remote Dev Machine) with these variables pre-configured.
41
42
 
43
+ ## Backend API access
44
+
45
+ Backend calls go through the evalforge ambassador packages (gRPC via
46
+ `@wix/http-client`): all reads, plus `addResult` (`AddEvalRunResult`),
47
+ `clearResults` (`ClearEvalRunResults`), and trace-event push
48
+ (`PushTraceEvent`). The only call still on the legacy REST surface is
49
+ `updateEvalRun` — the gRPC `UpdateEvalRun` handler only forwards
50
+ user-editable fields, not the system state transitions
51
+ (`status`/`completedAt`/`jobError`/`jobStatus`) the evaluator writes.
52
+
42
53
  ## Scripts
43
54
 
44
55
  ```bash