feat(evals): add Tone Appropriateness onto the ela-writing taxonomy

[adnanrhussain]

Migrates Tone Appropriateness onto the evals/feedback/ela-writing/ taxonomy as Tone Appropriateness, at evals/feedback/ela-writing/tone-appropriateness/. Same pattern as the student-facing-text family (#159, #161, #163, #173–#177). The evaluator already had a complete contract; this moves it, renames it, and closes the gaps that family standardised on.

Evaluator

• evaluator.id: feedback.ela_writing.tone_appropriateness
• stable_id: e6d674bd-2c0c-4f8d-9582-e9924c39445f, id_history: ["feedback.productive_coaching_writing_feedback.is_tone_appropriate"]
• evaluator.name: Tone Appropriateness Evaluator — drops the Feedback-Quality suffix so it matches the <X> Evaluator form used across student-facing-text
• Step id evaluate_is_tone_appropriate → evaluate_tone_appropriateness
• Single LLM step on gpt-5.4-2026-03-05, temperature 1, unchanged

quality_score

The output score field was named after the evaluator (tone_appropriateness_score). Every evaluator in this family had a differently-named field for the same binary judgement, so a consumer had to know which evaluator ran before it could read the result.

Student-facing-text names this after the measured dimension — complexity_score, identical across all 8 of its evaluators. This adopts the same convention: quality_score, shared across all 7 feedback evaluators. Values are unchanged (integer, 0/1).

Renamed in output_schema.json (property, required[], description), fixtures.json (expected), and the notebook. Nothing consumes these configs — no TypeScript or Python SDK reads them — so there is no downstream breakage.

Also fixed

The notebook carried two copy-paste references to evals/prompts/purpose, a leftover from the Purpose evaluator it was templated from. Both now point at this directory.

Not applicable

The grade_level string standardisation doesn't apply — this evaluator takes student_text and feedback_text, no grade input. supported_grades: ["8", "9"] is metadata only.

Verification

scripts/check.py all pass.

[Copilot]

Pull request overview

Adds the Tone Appropriateness feedback evaluator into the evals/feedback/ela-writing/ taxonomy and aligns it with the shared config-driven evaluator contract used across related evaluators.

Changes:

• Introduces canonical prompt assets (system.txt, user.txt) plus a new input_schema.json for the evaluator.
• Standardizes the binary output field name to quality_score across schema, fixtures, and the example notebook.
• Updates evaluator identity metadata (evaluator.id, stable_id, id_history) and renames the step id to evaluate_tone_appropriateness.

Reviewed changes

Copilot reviewed 4 out of 7 changed files in this pull request and generated no comments.

Show a summary per file

File	Description
evals/feedback/ela-writing/tone-appropriateness/user.txt	Adds the user prompt template with {student_text} / {feedback_text} placeholders.
evals/feedback/ela-writing/tone-appropriateness/system.txt	Adds the system prompt rubric/instructions for tone appropriateness.
evals/feedback/ela-writing/tone-appropriateness/output_schema.json	Renames the output score field to quality_score and updates schema description.
evals/feedback/ela-writing/tone-appropriateness/input_schema.json	Adds an input JSON schema for student_text and feedback_text.
evals/feedback/ela-writing/tone-appropriateness/fixtures.json	Updates fixtures to expect quality_score.
evals/feedback/ela-writing/tone-appropriateness/example_notebook.ipynb	Updates notebook narrative + helper function + fixture evaluation to use quality_score and new naming.
evals/feedback/ela-writing/tone-appropriateness/config.json	Updates evaluator identifiers/metadata and step id; pins prompt hashes and schema refs.

---

💡 Add a code-review agent skill or configure MCP servers for context-aware, tailored reviews. Learn more in the docs.