free-recall explain tool: the quizzing-mechanic switch (owner 2026-09-27)

Sycophancy mitigation as a structural commitment: the model writes its
full correctAnswer (plus explanation, in that order) BEFORE the learner
sees the question; the commitment rides the persisted call args, is
revealed on submit, and grades his free text in the model's next reply —
the client never grades free recall.

- tools.explain setting (segmented 'Quizzing — Multiple choice | Free
  recall' in Agent settings' Tools group): free arms explain and drops
  quiz for BOTH agents (prober variant carries no nodeId, same pattern
  as the probe quiz); mechanics keys never pass filterEnabledTools —
  quiz stays unwritable (the 2026-09-19 wedge ruling stands)
- doctrine lives pi-natively in the tool descriptions (owner ruling —
  no conditional system-prompt chunks): explain carries the commitment
  rule; the MC option-building craft moved out of the prompt into the
  quiz description; shared/20-question-tools.md slims to the
  quiz-agnostic core
- display seams: parseExplainInput (LaTeX-debris restoration included),
  dock text-mode with an 'I don't know' row (kbd row 0), AnsweredCard
  reveal (The answer / Why, no verdict chips), [Explain answer] format
- phases/plan seams recognize both quizzing tools' nodeId tags; explain
  answers deliberately feed no ledger evidence (no client grading to
  key on — a future ruling can parse the model's grade)

1049/1049 tests, typecheck/lint/build green.
This commit is contained in:
amosblomqvist
2026-09-27 08:55:13 +03:00
parent 07d2c84461
commit 2f39766528
25 changed files with 784 additions and 68 deletions
+8 -1
View File
@@ -39,7 +39,7 @@ export interface PendingToolCall {
}
/** Tools that render in the question dock and gate the composer. */
export const QUESTION_TOOLS = ["quiz", "ask_user_question"] as const;
export const QUESTION_TOOLS = ["quiz", "ask_user_question", "explain"] as const;
/** Tools fulfilled by background sub-agent jobs (W2/W6). */
export const JOB_TOOLS = ["research", "demo", "visual"] as const;
@@ -62,6 +62,13 @@ export const JOB_TOOLS = ["research", "demo", "visual"] as const;
* 2026-09). */
export const TOGGLEABLE_TOOLS = ["research", "demo", "visual", "pdf"] as const;
/** The keys the settings WRITE side accepts in agent.tools: the plain
* toggleables PLUS the quizzing-mechanics switch (`explain: true` arms
* free-recall quizzing and drops the MC quiz — see quizMechanics in
* agent/tools.ts; owner 2026-09-27). `quiz` itself stays unwritable:
* mechanics-managed, never individually toggleable. */
export const WRITABLE_TOOL_KEYS = [...TOGGLEABLE_TOOLS, "explain"] as const;
/** The resolved settings → factory seam shape (Platform wires chat-handler
* to pass this through from the settings record; storage and defaults are
* Platform's lane). JSON-native: a partial record keyed by tool name.