docs(migration): align skill and trajectory assertion docs (#1687)

* docs(migration): align skill and trajectory assertions

* docs: canonicalize trajectory assertion examples

Author: christso

apps/web/src/content/docs/docs/next/evaluation/batch-cli.mdx

@@ -111,9 +111,9 @@ JSONL where each line is a JSON object with an `id` matching a test:
 
 The `id` field must match the test `id` for AgentV to route output to the correct grader.
 
-### Output with Tool Trajectory
+### Output with Trajectory Assertions
 
-To enable `tool-trajectory` evaluation, include `output` with `tool_calls`:
+To enable `trajectory:*` assertions, include `output` with `tool_calls`:
 
 ```json
 {
@@ -138,7 +138,7 @@ To enable `tool-trajectory` evaluation, include `output` with `tool_calls`:
 }
 ```
 
-AgentV extracts tool calls directly from `output[].tool_calls[]` for `tool-trajectory` graders.
+AgentV extracts tool calls directly from `output[].tool_calls[]` for trajectory assertions.
 
 ## Grader Implementation
 

apps/web/src/content/docs/docs/next/evaluation/examples.mdx

@@ -124,7 +124,7 @@ tests:
 
 See [`examples/features/file-transforms/`](../../../../../examples/features/file-transforms/) for a runnable end-to-end example with a file-producing target and custom grader target.
 
-## Tool Trajectory
+## Trajectory Assertions
 
 Validate that an agent uses specific tools during execution:
 
@@ -139,23 +139,27 @@ tests:
     input: Research REST vs GraphQL
     assert:
       - name: research-check
-        type: tool-trajectory
-        mode: any_order
-        minimums:
-          knowledgeSearch: 2
-          documentRetrieve: 1
+        type: trajectory:tool-used
+        value:
+          name: knowledgeSearch
+          min: 2
+      - type: trajectory:tool-used
+        value:
+          name: documentRetrieve
+          min: 1
 
   # Validate exact tool sequence
   - id: auth-flow
     criteria: Agent follows auth sequence
     input: Authenticate user
     assert:
       - name: auth-sequence
-        type: tool-trajectory
-        mode: exact
-        expected:
-          - tool: checkCredentials
-          - tool: generateToken
+        type: trajectory:tool-sequence
+        value:
+          mode: exact
+          steps:
+            - checkCredentials
+            - generateToken
 ```
 
 ## Offline Grader Benchmark
@@ -204,11 +208,12 @@ tests:
     input: Analyze trace
     assert:
       - name: trace-check
-        type: tool-trajectory
-        mode: in_order
-        expected:
-          - tool: webSearch
-          - tool: readFile
+        type: trajectory:tool-sequence
+        value:
+          mode: in_order
+          steps:
+            - webSearch
+            - readFile
 ```
 
 ## Multi-Turn Conversation

apps/web/src/content/docs/docs/next/evaluation/rubrics.mdx

@@ -199,7 +199,7 @@ assert:
   - The agent used Read to inspect the config file before editing
 ```
 
-This is a lightweight alternative to the `skill-trigger` evaluator when you want to check tool usage with natural-language criteria.
+Use this for qualitative tool-use checks. Use `skill-used`, `not-skill-used`, or `trajectory:*` assertions when the requirement can be deterministic.
 
 ## Combining with Other Graders