Manage prompts and conversations by using AI

AB-730 · 80 questions

• A permitting clerk writes “help with this” versus stating the goal, audience, and tone for a deficiency letter. What makes the second approach more effective?
• A 311 lead improves Copilot answers by stating whether the output is for email or a phone script. What prompt practice does this demonstrate?
• A parks educator adds grade-level reading constraints when prompting Copilot for trail signs. Why does this improve effectiveness?
• A budget analyst includes the decision question and explicit non-goals so Copilot does not invent recommendations. What prompt skill is this?
• A housing caseworker prompts for bullet options first, then asks for a full letter. What prompting approach is this?
• A clerk specifies “use only the attached fee schedule; say if missing” when prompting Copilot. What does this grounding instruction help reduce?
• An emergency planner includes time horizon and severity tags when asking Copilot to outline a briefing. What does this add to the prompt?
• A library marketer avoids a one-word prompt and instead states purpose, brand voice, and call to action. What contrast does this illustrate?
• A zoning reviewer attaches the adopted ordinance PDF rather than an outdated draft when prompting Copilot. What resource-selection practice is this?
• A grants team references the NOFO and the city’s prior narrative instead of random web blogs. What selection principle are they applying?
• A transit planner points Copilot at the approved ridership workbook instead of a personal scratch sheet. Why is that the better reference?
• A public-health writer chooses the current vaccine FAQ page over last year’s Word draft. What resource quality are they prioritizing?
• A courts clerk declines to reference a sealed personnel file when drafting a public hearing notice. What resource judgment does this show?
• A utilities lead selects both an outage map export and the customer-comms template as Copilot references. What selection approach is this?
• A permitting supervisor saves a well-tested deficiency-letter prompt for reuse each week. What Copilot practice does this demonstrate?
• A 311 trainer runs a de-escalation rewrite prompt several times until it consistently matches city tone rules. What should the trainer do next to make that prompt easy to reuse?
• A parks coordinator expects a summer hiring rush of seasonal event announcements. How should the coordinator prepare Copilot for that recurring work?
• Budget analysts rewrite the same capital improvement plan (CIP) narrative instructions from scratch each cycle. What Copilot prompt action best reduces that wasted effort?
• City fee-schedule language changes after council adoption. A clerk’s saved Copilot prompt still cites the old wording. What should the clerk do?
• A sanitation analyst needs a route-exception digest drafted every Monday without manually running the same Copilot request. What should the analyst do?
• During wildfire season only, an emergency-management lead wants a morning status draft from Copilot each day. Which action fits that need?
• A library director wants weekend program stats summarized before each Monday staff meeting. How should Copilot be set up?
• A transit ops planner has a one-time corridor briefing prompt that will not repeat on a calendar. What is the best choice about scheduling?
• A housing supervisor wants open work-order themes gathered automatically for each Friday stand-up. What Copilot action supports that ritual?
• A permitting manager wants counter staff to use the same deficiency-letter wording. What should the manager do with the saved Copilot prompt?
• A city communications lead needs PIOs in multiple departments to use the same approved crisis-tone guidance in Copilot. What action best supports that?
• A clerk has a prompt that helps inspectors phrase internal inspection notes. Who is the most appropriate audience when sharing that prompt?
• Seasonal parks supervisors need the same volunteer-onboarding structure in Copilot each year. What should the parks lead do?
• A grants writer needs last month’s Copilot chat that drafted the outcomes section. What should the writer do first?
• Before a zoning hearing, an analyst wants to reopen an earlier Copilot chat that compared two ordinance options. What skill does that require?
• A 311 supervisor wants to reuse a Copilot explanation that successfully resolved a billing-dispute pattern. How should the supervisor locate that wording?
• An emergency planner has many Copilot chats and needs the one that outlined shelter messaging before updating it. What should the planner do?
• A library staffer forgot the title of a Copilot chat that held program-copy drafts. How can the staffer recover that thread?
• A clerk accidentally pasted a resident phone list into a Copilot chat while drafting. What conversation-management action best addresses the leftover sensitive content?
• Before a records review, a parks intern has many exploratory Copilot chats that are no longer needed. What should the intern do?
• A budget analyst created a duplicate Copilot chat that was only a failed prompt experiment. What is an appropriate next step?
• A housing worker exported the final approved resident letter into the case file and no longer needs the Copilot drafting chat. What action fits that lifecycle?
• A city attorney’s aide mixed privileged notes into a Copilot chat used for public-hearing prep, then split the work into separate threads. What should happen to the mixed chat?
• A permitting lead’s Copilot thread is still titled "New chat" but covers third-quarter porch-permit FAQs. What should the lead do to make it easier to find later?
• A transit planner wants a Copilot chat about a corridor briefing to stay identifiable months later. What rename practice helps most?
• A public-health writer juggles Copilot chats for a heat-advisory campaign and a vaccine-clinic campaign. What rename practice best prevents mix-ups?
• A library marketer’s Copilot chat still uses last month’s program title after the program is renamed mid-drafting. What should the marketer do?
• An emergency ops specialist wants overnight shift staff to recognize which Copilot chat holds the active Incident Action Plan language. What helps most?
• A capital improvement program analyst has a useful Copilot conversation about bond-project research. How should the analyst gather that chat with related notes?
• A grants team wants one Copilot chat parked with other source notes for a single funding application. What action fits?
• A zoning researcher ran several ordinance-comparison Copilot chats and needs them together for a hearing packet. What should the researcher do?
• A parks planner wants a trail-plan brainstorm Copilot chat available as part of an ongoing project working set. What is the best next step?
• A 311 quality lead wants exemplary Copilot chats collected for new-agent training. How should those chats be organized?
• A city clerk needs a records-request helper and sees a suitable option already listed in Agent Store. What should the clerk do first?
• A utilities lead reviews Agent Store and finds no agent that can use the city’s outage-communications knowledge pack. What is the appropriate choice?
• A library director needs reading-program insights and sees specialized Researcher or Analyst-style agents in Agent Store. What judgment fits AB-730 practice?
• A housing team discovers Agent Store already offers an agent that drafts plain-language resident letters. What should the team avoid?
• A permitting office confirms that Agent Store agents cannot ground on the city’s local fee PDFs. What decision follows?
• A city IT champion teaches staff how to approach Copilot agents for municipal tasks. Which guidance matches store-first judgment?
• A parks supervisor needs a new Copilot agent and does not want to configure every field from a blank form. What should the supervisor do?
• A 311 manager is creating a resident-response agent and must choose a starting point in the create flow. Which choice is best?
• A grants office wants to accelerate agent setup for NOFO workflows without keeping a generic research template unchanged. What approach fits?
• A clerk wants an agent that already includes a structure for suggested prompts covering common counter questions. Why create from a template?
• An emergency planner needs instructions and capabilities scaffolding present when creating a Copilot agent. What should the planner use?
• A transit communications lead sees a close template in the agent create flow and also sees a blank custom option. What should the lead choose?
• A permitting team is configuring a Copilot agent that must answer from official fee and zoning guidance. What knowledge action is required?
• A public-health team wants a Copilot agent to follow approved clinic SOPs and avoid draft research notes. How should knowledge be scoped?
• A housing agent keeps missing the city’s resident-letter style until a caseworker updates configuration. What change enables grounded answers?
• A libraries agent invents program dates instead of using the real calendar. What knowledge fix should staff apply?
• Utilities revises the outage-communications playbook that a Copilot agent uses. What should staff do to keep answers current?
• A courts clerk finds the department agent still cites an obsolete policy PDF. What knowledge change should the clerk make?
• A CIP planning agent must stay trustworthy for council briefings. How should knowledge be bounded?
• A 311 supervisor configures a resident-help agent. Which instructions setting best fits the desk?
• A parks events agent should summarize approved event PDFs without wandering the open web. What capability choice is appropriate?
• A permitting desk wants counter staff to start the fee-FAQ agent quickly. What settings action helps most?
• A housing authority wants its casework agent to show where answers came from. Which instructions change supports that?
• An emergency-management coordinator configures suggested prompts for a field agent. Which set best matches operations?
• A library children’s-program agent started drafting content outside its scope. What should the librarian adjust?
• A grants office agent sometimes invents match percentages missing from source files. Which instructions guardrail helps?
• A transit delay-notice agent needs a complete settings pass before launch. What bundle is most complete?
• A permitting supervisor finishes a custom fee-FAQ agent for the counter. What sharing action should come next?
• A public-health lead builds an outbreak-comms agent for PIOs. How should sharing be scoped?
• Seasonal parks leads need the same volunteer-briefing helper. What should the parks manager do?
• A housing director shares a casework agent and wants proof teammates can use it. What confirmation matters most?
• City IT finds staff copying a sensitive casework agent’s prompts into personal notebooks instead of using the approved share. What guidance is best?