docs(dashboard-agent): make the eval docs match the code

The docs claimed every turn is judged; the sampled rate, the redaction rule, the code-mode
skip, the opt-out and the retention period are now written down in one place.

Author: kathiekiwi

internal-packages/dashboard-agent/GUIDEBOOK.md

@@ -479,5 +479,7 @@ left standing ages into its own baseline and starts reading as normal.
 - Invent numbers or claim something doesn't exist beyond a truncated page.
 - Trust a report whose telemetry is stale.
 
-Feedback → #dashboard-agent-feedback, or just tell the agent — every
-conversation is evaluated and capability gaps are collected automatically.
+Feedback → #dashboard-agent-feedback, or just tell the agent — a sample of turns
+is scored automatically and capability gaps are collected from it (a tenth by
+default, none for a turn that read source, none for an org that opted out; see
+[README.md](./README.md#turn-evals)).

internal-packages/dashboard-agent/README.md

@@ -43,3 +43,23 @@ import type { dashboardAgent } from "@internal/dashboard-agent";
 ```
 
 Never a value import (see `src/index.ts`).
+
+## Turn evals
+
+A sampled fraction of turns is scored by an LLM judge (`dashboard-agent-eval-turn`), which
+writes one `chat_turn_evals` row. The rules live in one file, `src/eval-policy.ts`:
+
+- **Sampling.** `DASHBOARD_AGENT_EVAL_SAMPLE_RATE`, default **0.1** — the judge is a full
+  model call per turn and nothing reads the rows yet. Golden / CI runs are a separate lane:
+  `DASHBOARD_AGENT_EVAL_SAMPLE_RATE_CI` (default 1) applies only when
+  `DASHBOARD_AGENT_EVAL_CONTEXT=ci`, so neither lane can change the other's rate.
+- **Redaction.** Run payloads and outputs, query result rows, file contents and span
+  attributes are replaced by their shape before the turn leaves the agent. The row keeps the
+  judge's derived verdict only — never the question, the answer, or any tool data.
+- **Code mode.** A turn that called a source tool is not judged at all.
+- **Opt-out.** Per-org, via the `dashboardAgentTurnEvalsEnabled` feature flag. The agent asks
+  the API before every judged turn and judges only on an explicit yes.
+- **Retention.** Rows are dropped after 30 days by the webapp's dashboard-agent sweep.
+
+When a document and the code disagree about any of the above, the code is the fact and the
+document is the bug: fix the document in the same change.