practice: target-validation pins + the id convention in the base prompt (round 9 slice 3)

Four pins on the mastery-trace contract: a name target normalizes to
the canonical id (case-insensitive, whole-graph set), an id passes
through untouched, an unknown target rejects one-line AND does not arm
the gate (honest activation — the next valid pose goes through), and
the legacy node-source-less wiring stays unvalidated. The base prompt
now states the convention: target takes the proposition ids from the
session context — ids are how problems are traced.

src/agent/prompts/practice/00-base.md

@@ -23,7 +23,7 @@ He has to be able to trust you completely; one confidently-delivered hallucinati
 
 Two tools, and everything else is conversation. The mechanics only — the method (when to pose, what to pose, how to respond) is the doctrine's.
 
-**`problem`** — poses one problem. The statement field is what the learner sees (markdown and LaTeX render); the modelSolution is your private commitment, hidden from the learner until you conclude. The call ends the turn. The learner then works in a thinking field; each submission arrives back as a *Your thinking* message. Whether it claims an answer is IN THE TEXT, not declared anywhere — read it:
+**`problem`** — poses one problem. The statement field is what the learner sees (markdown and LaTeX render); the modelSolution is your private commitment, hidden from the learner until you conclude. The target field takes the proposition ids from the session context — ids are how every problem is traced to the propositions it exercises. The call ends the turn. The learner then works in a thinking field; each submission arrives back as a *Your thinking* message. Whether it claims an answer is IN THE TEXT, not declared anywhere — read it:
 
 - Working trace (mid-work, stuck, or asking for a nudge): the hint ladder applies, one rung per request. A trace may span several submissions.
 - A claimed answer: extract the moves, deliver the verdict. A claim is not a progress signal.

tests/practice-agents.test.ts

@@ -135,6 +135,50 @@ describe("the practice tool gating law", () => {
 	});
 });
 
+// ── target validation (the mastery-trace contract, owner round 9) ──────────
+
+describe("target validation — names normalize to ids, unknowns reject", () => {
+	const NODES = [
+		{ id: "n_parity", name: "Parity of a sum", statement: "The sum of two odds is even." },
+		{ id: "n_pairing", name: "Pairing", statement: "Pair elements to count parity." },
+	] as const;
+	const graphNodes = () => NODES;
+
+	it("a NAME target silently normalizes to the canonical id (the record carries ids)", async () => {
+		const healed = problemTool(createPracticeGate(ctxOf([])), graphNodes).prepareArguments!({
+			...PROBLEM_INPUT,
+			target: ["Parity of a sum", "pairing"],
+		}) as { target: string[] };
+		assert.deepEqual(healed.target, ["n_parity", "n_pairing"], "case-insensitive name → id, whole-graph set");
+	});
+
+	it("an id target passes through untouched (already canonical)", async () => {
+		const healed = problemTool(createPracticeGate(ctxOf([])), graphNodes).prepareArguments!({
+			...PROBLEM_INPUT,
+			target: ["n_parity"],
+		}) as { target: string[] };
+		assert.deepEqual(healed.target, ["n_parity"]);
+	});
+
+	it("an UNKNOWN target rejects with a one-line error and does NOT arm the gate", async () => {
+		const gate = createPracticeGate(ctxOf([]));
+		const tool = problemTool(gate, graphNodes);
+		const rejected = await tool.execute("call_x", { ...PROBLEM_INPUT, target: ["ghost-prop"] } as never);
+		assert.equal(rejected.isError, true);
+		assert.match((rejected.details as { problems: string[] }).problems[0]!, /ghost-prop/);
+		assert.match((rejected.details as { problems: string[] }).problems[0]!, /proposition ids from the session context/);
+		// honest activation (the BKT law): a rejected pose frees the slot —
+		// the next, valid pose goes through.
+		const valid = await tool.execute("call_y", { ...PROBLEM_INPUT, target: ["n_parity"] } as never);
+		assert.equal(valid.isError, undefined);
+	});
+
+	it("NO node source (the legacy/test wiring) = no validation, free strings ack", async () => {
+		const result = await problemTool(createPracticeGate(ctxOf([]))).execute("call_x", PROBLEM_INPUT as never);
+		assert.equal(result.isError, undefined);
+	});
+});
+
 // ── the registry + the agent ────────────────────────────────────────────────
 
 describe("the practice agent", () => {