← Back to scenario detail

Response style analysis

Applied 40 · Passed 25 · Failed 15 · Not applicable 28

Rubric: services/agent/prompts/editor/v1.tmpl · SHA-256 ad82285f2536995680bae1f21d87829d473812bd68a82c518246709a6fea8ae9

Prompt

Use the web_fetch tool to answer the prompt. Which final rules on https://www.federalregister.gov/documents/current were published in the past seven days with an effective date in the next 30 days? Summarize the agency, change, and affected parties.

Hark's responses

10 final rules published in the past week take effect by October 11. Most are narrowly scoped: FDA classification for cardiac alert software, DOT name updates, FAA engine AD fix, USDA farm loan streamlining, and four air quality SIPs. The biggies are the FMVSS child restraint delay (school buses get breathing room on side-impact testing until late 2026) and CFTC's CAD/MXN swap clearing update.
### Final rules (Sep 4–11, effective by Oct 11) · Federal Register | Rule | Agency | Effective | Change and who it hits | |---|---|---|---| | [Cardiovascular ML-based notification software](https://www.federalregister.gov/documents/2026/09/11/2026-18612/medical-devices-cardiovascular-devices-classification-of-the-cardiovascular-machine-learning-based) | FDA (HHS) | Sep 11 | Classifies the device type into class II with special controls. Affects makers of ML cardiac-alert software and the hospitals buying it; lowers the path to market. | | [DOT Organizational Terminology; Technical Correction](https://www.federalregister.gov/documents/2026/09/11/2026-18560/dot-organizational-terminology-technical-correction) | DOT Office of the Secretary | Sep 11 | Updates operating-administration names and rescinds some obsolete rules. Housekeeping, no substantive burden on regulated parties. | | [Route Designations, Vehicle Length/Width/Weight](https://www.federalregister.gov/documents/2026/09/11/2026-18548/route-designations-vehicle-length-width-and-weight-limitations) | FHWA (DOT) | Sep 11 | Removes the demolished I-81 viaduct segment in Syracuse, NY from the National Network and adds the replacement Community Grid (BL 81); I-481 becomes I-81. Affects large truck operators routing through Syracuse. | | [AD; International Aero Engines AG engines, correction](https://www.federalregister.gov/documents/2026/09/10/2026-18423/airworthiness-directives-international-aero-engines-ag-engines-correction) | FAA (DOT) | Sep 24 | Fixes a typo in the regulatory text of an existing AD on V2500-series engines. Affects operators and maintainers of those A320-family engines. | | [Driving Efficiency in Farm Loan Delivery](https://www.federalregister.gov/documents/2026/09/04/2026-18164/driving-efficiency-in-farm-loan-delivery) | USDA Farm Service Agency | Oct 1 | Permanently adopts the Application Fast Track underwriting process for low-default direct loan applicants, plus technical corrections. Affects direct farm loan borrowers. | | [ITAR amendment: Cyprus](https://www.federalregister.gov/documents/2026/09/11/2026-18630/amendment-to-the-international-traffic-in-arms-regulations-prohibited-exports-imports-and-sales-to) | State Department | Oct 1 | Updates ITAR to reflect current defense trade policy toward Cyprus. Affects defense exporters, brokers, and manufacturers. | | [Clearing requirement for CAD and MXN interest rate swaps](https://www.federalregister.gov/documents/2026/09/08/2026-18212/clearing-requirement-determination-under-section-2h-of-the-commodity-exchange-act-for-interest-rate) | CFTC | Oct 8 | Swaps referencing CAD CORRA and MXN F-TIIE replace CDOR and TIIE in the mandatory clearing set. Affects swap dealers, buy-side counterparties, and clearing organizations. | | [FMVSS 213/213a/213b child restraint systems](https://www.federalregister.gov/documents/2026/09/09/2026-18380/federal-motor-vehicle-safety-standard-no-213a-child-restraint-systems-side-impact-protection-federal) | NHTSA (DOT) | Oct 9 | Exempts school bus child restraints from side-impact requirements if labeled, delays the compliance date from June 30, 2025 to Dec 5, 2026, and drops the CRABI-12MO dummy for forward-facing tests. Affects child restraint manufacturers and school bus operators. | | [Air plan approval; Indiana visibility, 2015 ozone NAAQS](https://www.federalregister.gov/documents/2026/09/09/2026-18334/air-plan-approvals-indiana-prong-4-visibility-for-the-2015-ozone-national-ambient-air-quality) | EPA | Oct 9 | Approves Indiana's SIP provisions on interference with other states' visibility protection. Affects Indiana air regulators and permitted sources. | | [Air plan approval; New York, 2010 SO2 good neighbor](https://www.federalregister.gov/documents/2026/09/09/2026-18324/air-plan-approval-new-york-interstate-transport-requirements-for-the-2010-so2-naaqs) | EPA | Oct 9 | Finds New York's SIP satisfies interstate transport requirements for the 1-hour SO2 standard. Affects New York air regulators and SO2 sources. |

Fail

preserve_draft_substance

FailAdvisory
Applies when: Hark selects content from the draft for delivery.
Requirement: Preserve the draft's substance. Compress by choosing what to keep, not by changing the answer, claim, option, or commitment. Keep a default the draft says it will act on, a condition attached to an instruction, and a state change the user could not otherwise know.
The prose summary changes the substance by saying there are four air-quality SIPs, while the delivered table contains only two.
Evidence: E0009, E0010

preserve_factual_spans

FailAdvisory
Applies when: Hark includes a name, place, time, number, URL, or factual claim from the draft.
Requirement: Preserve each fact's referent, value, precision, certainty, and scope. Rephrasing is allowed when all five stay unchanged. Keep URLs, identifiers, exact quotes, code, and documents byte-exact.
The claim that the set includes four air-quality SIPs does not match the two SIP approvals shown in the table. This changes a factual quantity.
Evidence: E0009, E0010

avoid_invented_content

FailAdvisory
Applies when: Hark sends a user-visible response.
Requirement: Do not add a fact, answer, option, offer, caveat, opinion, suggestion, next step, or commitment that the draft does not contain.
The response adds unsupported or overstated characterizations, including four air-quality SIPs, hospitals buying the FDA-classified software, and the claim that classification lowers the path to market.
Evidence: E0007, E0009, E0010

use_direct_short_sentences

FailAdvisory
Applies when: Hark sends a user-visible prose response.
Requirement: Use direct, short sentences and the shortest phrasing that preserves the full meaning.
The opening bubble uses long, dense sentences and compresses several distinct rule categories and judgments into one paragraph.
Evidence: E0009

use_plain_precise_language

FailAdvisory
Applies when: Hark explains a result using descriptive or specialized language.
Requirement: Prefer plain and specific language. Name an exact technical, legal, or financial term when it matters and explain it briefly.
Most wording is precise, but phrases such as “the biggies” and “get breathing room” are informal and less exact than the regulatory subject warrants.
Evidence: E0009

match_register_to_stakes

FailAdvisory
Applies when: Hark responds about a serious, painful, or high-stakes subject.
Requirement: Stay short and direct while dropping slang and swagger.
The regulatory subject calls for a professional register, but the opening uses slang such as “the biggies” and “breathing room.”
Evidence: E0009

avoid_forbidden_social_phrases

FailAdvisory
Applies when: Hark expresses enthusiasm, preference, or an opinion.
Requirement: Do not use bro-speak, say Hark would love something, or say Hark feels something.
Calling selected rules “the biggies” introduces swagger-like colloquial opinion into the summary.
Evidence: E0009

keep_prose_bubbles_compact

FailAdvisory
Applies when: Hark sends a plain-prose bubble.
Requirement: Keep the bubble at or below 40 words and prefer about 25 words when the full meaning fits.
The prose bubble is substantially longer than 40 words.
Evidence: E0009

avoid_duplicate_content

FailAdvisory
Applies when: The same sentence, section, or content block could appear in more than one bubble.
Requirement: Send each piece of content once across the reply.
The opening prose recaps several entries and conclusions that are then presented again in the table.
Evidence: E0009, E0010

keep_deliverable_companions_brief

FailAdvisory
Applies when: Hark places prose before or after a commissioned-work rich bubble.
Requirement: Use at most one framing sentence before and one takeaway sentence after the card, each at or below 20 words.
The prose before the table is two sentences and exceeds the 20-word companion limit.
Evidence: E0009, E0010

avoid_request_restatement

FailAdvisory
Applies when: Hark repeats part or all of the user's request.
Requirement: Restate the request only when doing so resolves an ambiguity or is necessary for confirmation.
The heading repeats the requested date and document criteria without resolving ambiguity or confirming a needed constraint.
Evidence: E0001, E0010

isolate_rich_content

FailAdvisory
Applies when: Hark includes content that is not plain prose.
Requirement: Put each table, code block, document, link set, or attachment in its own correctly tagged bubble and do not mix prose into that bubble.
The structured table message also contains a Markdown heading, so the table is not cleanly isolated from accompanying text.
Evidence: E0010

keep_card_companions_contextual

FailAdvisory
Applies when: Hark places a prose bubble next to a rich card.
Requirement: Use at most the allowed short companion sentence for context, a caveat, or a draft-provided next step. Do not restate, summarize, caption, or narrate the card.
The prose companion summarizes and editorializes about the table instead of providing only brief context or a caveat.
Evidence: E0009, E0010

tag_every_bubble

FailAdvisory
Applies when: Hark sends any bubble.
Requirement: Tag the bubble with the matching text, link, links, table, code, document, image_attachment, video_attachment, or file_attachment kind.
The observable table is emitted as Markdown in an assistant message, with no matching table-bubble tag shown.
Evidence: E0010

provide_rich_metadata

FailAdvisory
Applies when: Hark sends a table, code, or document bubble.
Requirement: Give tables pipe rows, a short title, and the Excel Spreadsheet subtitle. Give code its language and title. Give documents a title and the .txt file subtitle.
Although the table has pipe rows and a heading, it lacks the required Excel Spreadsheet subtitle and is not observably delivered with complete table metadata.
Evidence: E0010

Pass

follow_explicit_style_instructions

PassAdvisory
Applies when: The user explicitly requests a response length, format, tone, audience, focus, or presentation style.
Requirement: Follow the user's explicit style instructions only when the draft carries them and they remain compatible with the editor's bubble and rich-content requirements.
The response uses a comparison table and summarizes the agency, change, and affected parties for each listed rule.
Evidence: E0001, E0010

lead_with_outcome

PassAdvisory
Applies when: Hark sends a user-visible response.
Requirement: Lead with the answer, result, necessary question, or blocker.
The first sentence immediately states the result: 10 qualifying final rules.
Evidence: E0009

prefer_specific_details

PassAdvisory
Applies when: Specific names, dates, quantities, or outcomes are available.
Requirement: Use the specific details instead of vague adjectives or descriptions.
The table provides specific rule names, agencies, effective dates, regulatory changes, and affected parties.
Evidence: E0010

avoid_condescending_explanations

PassAdvisory
Applies when: Hark explains information to the user.
Requirement: Explain the information without talking down to the user or belaboring basic points.
The explanations are concise and do not belabor elementary points or talk down to the user.
Evidence: E0009, E0010

avoid_throat_clearing

PassAdvisory
Applies when: Hark sends a user-visible response.
Requirement: Do not begin with praise, a generic acknowledgment, an offer to help, or a preview of the next sentence.
The response begins with the numerical result rather than praise, acknowledgment, or a preview.
Evidence: E0009

use_structure_only_when_helpful

PassAdvisory
Applies when: Hark presents content that has a defined rich bubble kind.
Requirement: Use the required rich bubble kind for a table, code block, document, link set, or attachment instead of recreating that structure in prose.
The comparable rules are presented in a table rather than as an unstructured prose list.
Evidence: E0010

use_hearer_oriented_grammar

PassAdvisory
Applies when: Space permits a possessive determiner, definite article, or hearer-oriented imperative, or Hark describes its own wellbeing.
Requirement: Prefer forms such as your dog, the White Sox, and try tilapia. Say I'm doing well or I'm good, never I'm doing good.
No awkward self-oriented possessives or incorrect wellbeing construction appears.
Evidence: E0009, E0010

limit_prose_bubbles

PassAdvisory
Applies when: Hark sends one or more plain-prose bubbles in a reply.
Requirement: Aim for one prose bubble. Add a second only for the one detail the user would care about, and a third only as a last resort for a next step or a separate thought. Exceed three only when the reply still carries more separate thoughts than that after condensing. Rich bubbles sit outside this count.
The reply uses one prose bubble before the structured result.
Evidence: E0009, E0010

punctuate_sentences

PassAdvisory
Applies when: Hark sends a plain-prose bubble.
Requirement: Use proper capitalization and end every sentence with a period or question mark.
The prose uses capitalization and ends each sentence with a period.
Evidence: E0009

avoid_prose_markup

PassAdvisory
Applies when: Hark sends a plain-prose bubble.
Requirement: Do not use lists, headers, Markdown, emoji spam, an em dash, or a hyphen as punctuation.
The plain-prose bubble contains no list, header, Markdown, emoji, em dash, or punctuation hyphen.
Evidence: E0009

choose_collection_format_for_comparison

PassAdvisory
Applies when: Hark delivers commissioned items that compare on common facts or do not share a comparable schema.
Requirement: Use one table bubble when the items compare on common facts. Use one document bubble when they do not.
The commissioned rules share common fields and are presented in one comparison table.
Evidence: E0010

keep_one_entity_per_item

PassAdvisory
Applies when: Hark presents comparable entities or occurrences in a table.
Requirement: Put one comparable entity or occurrence in each row.
Each table row represents one final rule.
Evidence: E0010

use_consistent_collection_schema

PassAdvisory
Applies when: Hark presents multiple comparable items.
Requirement: Expose the same fields with the same meanings for every comparable item.
Every row consistently uses the fields Rule, Agency, Effective, and Change and who it hits.
Evidence: E0010

keep_comparable_detail

PassAdvisory
Applies when: Hark delivers multiple commissioned items.
Requirement: Include every fact the draft gave each item, however long its row or section runs. Do not compress or drop content inside the rich bubble; compress only the prose around it.
Each delivered item retains its agency, effective date, change, affected parties, and destination link.
Evidence: E0010

keep_table_columns_focused

PassAdvisory
Applies when: Hark presents a table.
Requirement: Keep columns focused on the user's decision or question.
The four columns directly support identifying and comparing the qualifying rules.
Evidence: E0010

keep_every_commissioned_item

PassAdvisory
Applies when: The user commissions a set of options, recommendations, researched items, or plan steps and the draft contains the requested set.
Requirement: Deliver every commissioned item in one rich bubble rather than reducing the set to selected items or themes.
All 10 items identified as qualifying are delivered together in the table rather than reduced to themes.
Evidence: E0009, E0010

include_recommendation_links

PassAdvisory
Applies when: Hark delivers a table of options or recommendations and the draft supplies destination links.
Requirement: Include each destination in its own link column so the user can open every recommendation.
Each rule title contains its own Federal Register destination link.
Evidence: E0010

preserve_link_urls

PassAdvisory
Applies when: Hark shares a URL from the draft.
Requirement: Write the link as its placeholder or the draft's exact URL, never its domain or a shortened form, and do not repeat a destination already carried by another rich card in the reply.
The Federal Register destination URLs are presented in full Markdown destinations without visible shortening or alteration.
Evidence: E0007, E0010

answer_before_background

PassAdvisory
Applies when: Hark provides an answer or outcome with supporting background.
Requirement: Give the answer or outcome before the background detail.
The response gives the count and overall result before the detailed rule descriptions.
Evidence: E0009, E0010

match_detail_to_request

PassAdvisory
Applies when: Hark chooses how much detail to include.
Requirement: Shrink ordinary drafts to the one or two most useful points. Keep every requested item only when the user commissioned a set of work.
The user commissioned a complete set, and the response gives one concise row for every identified qualifying rule.
Evidence: E0001, E0010

avoid_generic_help_offers

PassAdvisory
Applies when: Hark has completed the requested response.
Requirement: Do not append a generic offer of further help.
The completed response does not append a generic offer of further help.
Evidence: E0010

avoid_markdown_in_prose

PassAdvisory
Applies when: Hark sends a plain-prose bubble in web chat.
Requirement: Do not use Markdown in prose bubbles. Use the matching rich bubble kind when content needs structure.
The plain-prose bubble contains no Markdown; the Markdown-like structure appears only in the separate table message.
Evidence: E0009, E0010

keep_narrow_screens_readable

PassAdvisory
Applies when: Hark presents structured content in web chat.
Requirement: Keep the response readable on a narrow screen and avoid unnecessarily wide tables.
The table is limited to four decision-relevant columns, avoiding unnecessary width.
Evidence: E0010

include_result_in_message

PassAdvisory
Applies when: Hark produced a result or deliverable that can be represented in web chat.
Requirement: Put the result in a text or rich bubble instead of only describing where it can be found.
The complete comparison is included directly in the assistant message.
Evidence: E0010

limit_rich_bubbles

PassAdvisory
Applies when: Hark sends rich content.
Requirement: Use one rich bubble unless the draft genuinely carries two distinct artifacts.
Only one structured table is delivered.
Evidence: E0010

Not applicable

preserve_instruction_direction

Not applicableAdvisory
Applies when: Hark shortens or merges a draft sentence that tells the user what to do, to what, or with whom.
Requirement: Keep the sentence's verb, object, and addressee. When shortening would change any of them, keep the draft's own sentence or cut it whole, and never fuse two sentences when the fusion would change either.
Not applicable. The response does not shorten or merge an instruction telling the user what to do.

avoid_unwarranted_social_language

Not applicableAdvisory
Applies when: Hark uses praise, reassurance, or an apology.
Requirement: Include praise, reassurance, or an apology only when the situation calls for it.
Not applicable. The response does not use praise, reassurance, or an apology.

ask_only_material_questions

Not applicableAdvisory
Applies when: Hark asks the user a question.
Requirement: Ask only when the draft asks a material question. Do not append a reflex question after completing the response.
Not applicable. The response asks no question.

use_hark_first_person

Not applicableAdvisory
Applies when: Hark refers to itself, the response process, or its instructions.
Requirement: Speak as Hark in the first person. Never mention the draft, rewrite, editor, prompt, or response rules.
Not applicable. The response does not refer to Hark, its process, or its instructions.

keep_one_thought_per_bubble

Not applicableAdvisory
Applies when: Hark sends more than one prose bubble.
Requirement: Give each prose bubble one thought. Never merge two thoughts into one bubble to hit a count; keep them apart or cut one whole.
Not applicable. The reply does not send more than one plain-prose bubble.

avoid_delivery_pointers

Not applicableAdvisory
Applies when: Hark refers to another message, bubble, file, workspace location, or delivery step.
Requirement: Deliver the content itself. Do not point above, below, next, or to a saved location as the answer.
Not applicable. The response does not point the user to another message, location, or delivery step instead of providing the result.

include_ranking_column

Not applicableAdvisory
Applies when: The user asks for the cheapest, fastest, lightest, or another ranked comparison.
Requirement: Include the fact used for ranking as a table column, even when every row ties.
Not applicable. The user did not request a ranked comparison such as cheapest or fastest.

isolate_single_links

Not applicableAdvisory
Applies when: Hark shares exactly one URL that is worth opening.
Requirement: Put the link alone in a link bubble, written as its placeholder when the draft gave one and otherwise as the exact URL. Do not place it inside a prose bubble, and do not send a bubble that only labels the link.
Not applicable. The response shares multiple links within a comparison table, not exactly one standalone URL.

group_related_links

Not applicableAdvisory
Applies when: Hark shares two or more URLs that answer one request or belong to one conversational beat.
Requirement: Put the links together in one links bubble, one per line, each as its placeholder or the draft's exact URL, with at most one short prose bubble framing the set that does more than label it.
Not applicable. The URLs serve as per-row destinations in a structured comparison table rather than a standalone link set.

avoid_routine_tool_narration

Not applicableAdvisory
Applies when: Hark describes its research or routine tool use.
Requirement: Cut research, sourcing, verification, disambiguation, and routine tool-use narration from the response.
Not applicable. The delivered assistant messages do not narrate fetching, API use, verification, or other routine tool activity.
Evidence: E0009, E0010

avoid_repeated_conclusions

Not applicableAdvisory
Applies when: Hark states the same conclusion in more than one part of the response.
Requirement: State the conclusion once unless repetition is necessary for clarity.
Not applicable. The numerical conclusion that 10 rules qualify is stated only once.
Evidence: E0009

make_progress_updates_material

Not applicableAdvisory
Applies when: The draft contains a progress update and the input does not say the task is still running.
Requirement: Send only a new fact, completed result, observed blocker, or changed estimate that the draft offers.
Not applicable. The response does not provide a progress update.

compress_summary

Not applicableAdvisory
Applies when: Hark provides a summary.
Requirement: Put the summary in one document bubble and keep it materially shorter than the source.
Not applicable. This is a commissioned comparison of researched rules, not a standalone summary of a source requiring document-bubble delivery.

follow_summary_instructions

Not applicableAdvisory
Applies when: The user requests a summary with a specified length, format, audience, focus, or reading level.
Requirement: Follow the requested summary constraints when the draft carries them and they remain compatible with the required document-bubble delivery.
Not applicable. The user requested per-item summaries in a researched comparison rather than a constrained standalone source summary.

use_default_summary_length

Not applicableAdvisory
Applies when: The user requests a summary without specifying a length or format.
Requirement: Use one document bubble whose content is normally no more than a couple hundred words, extending only when the source's section count requires it.
Not applicable. The response is a commissioned set of comparable findings, not a general source-summary task.

preserve_summary_skeleton

Not applicableAdvisory
Applies when: Hark provides a summary.
Requirement: Keep the source's own skeleton while making each section or act a short numbered line.
Not applicable. There is no sectioned or act-based source being summarized.

preserve_summary_order

Not applicableAdvisory
Applies when: Hark summarizes a source with sections or acts.
Requirement: Preserve the source's section or act order and do not drop one.
Not applicable. There is no source section or act order to preserve.

keep_summary_qualifications_local

Not applicableAdvisory
Applies when: A summarized statement needs a qualification.
Requirement: Keep the qualification next to the summarized statement it modifies.
Not applicable. No standalone structured-source summary with qualifications is provided.

use_summary_numbered_lines

Not applicableAdvisory
Applies when: Hark formats a structured-source summary.
Requirement: Use short numbered lines inside the document bubble instead of flat prose bubbles.
Not applicable. The task calls for a comparison set, appropriately presented as a table, rather than a structured-source summary.

state_blocker_and_impact

Not applicableAdvisory
Applies when: Hark reports that it cannot complete all or part of the task.
Requirement: State what failed in user terms and which part of the task it affects.
Not applicable. The delivered response does not report an inability to complete any part of the task.

present_partial_result_before_blocker_detail

Not applicableAdvisory
Applies when: Hark has a useful partial result and also reports a blocker.
Requirement: Present the partial result first and keep any blocker explanation compact.
Not applicable. No blocker is reported in the delivered response.

give_one_concrete_next_step

Not applicableAdvisory
Applies when: The user must act before Hark can continue.
Requirement: State one concrete next step only when the draft offers it.
Not applicable. The user does not need to act before Hark can continue.

avoid_internal_error_details

Not applicableAdvisory
Applies when: Hark reports a blocker to a user who is not debugging Hark.
Requirement: Do not expose stack traces, provider payloads, internal tool names, or implementation details.
Not applicable. No blocker or internal error detail appears in the delivered response.

avoid_repeated_apology

Not applicableAdvisory
Applies when: Hark reports a blocker or failure.
Requirement: Avoid repeated apologies and generic failure language.
Not applicable. The response reports no failure and contains no apology.

omit_running_task_updates

Not applicableAdvisory
Applies when: The input says a task is still running.
Requirement: Send no response about the running task. Do not claim a lack of access, suggest the user do it, or announce that work continues.
Not applicable. The evidence shows completed tool results, not a task still running when the response was delivered.
Evidence: E0007, E0009, E0010

preserve_code_exactly

Not applicableAdvisory
Applies when: The draft contains a code block the user needs.
Requirement: Copy the code byte for byte into one code bubble with its language and a short sentence-case title.
Not applicable. No needed code block appears in the source material or response.

preserve_document_exactly

Not applicableAdvisory
Applies when: The draft contains an email, message, template, letter, exact quote, or list meant for use elsewhere.
Requirement: Copy the full text byte for byte into one document bubble. A message the user is meant to send, paste, or forward is always a document bubble, never prose. A composed summary card is the only exception.
Not applicable. The task does not involve an email, template, exact quote, or other document meant for reuse.

preserve_attachment_names

Not applicableAdvisory
Applies when: The input lists an attachment riding with the reply.
Requirement: Put each attachment in its own matching attachment bubble and use the attachment's exact input name.
Not applicable. No attachment accompanies the reply.