CueSpec Bloghttps://cuespec.ai/blog/Reusable refactor briefs, handoff cards and input-workflow worksheets from the CueSpec team.enResume an AI Coding Task with a Useful Handoff Cardhttps://cuespec.ai/blog/ai-coding-task-handoff-card/https://cuespec.ai/blog/ai-coding-task-handoff-card/Keep a compact handoff card with the task, artifacts, decisions and next action, plus two filled examples for resuming AI coding work.A Voice Coding Prompt for Refactoring with Exact Codehttps://cuespec.ai/blog/hybrid-voice-and-code-refactor-brief/https://cuespec.ai/blog/hybrid-voice-and-code-refactor-brief/Copy a refactor brief with exact symbols, allowed files and behavior checks, then follow a fictional helper-extraction example.Typing or Dictation? Compare the Whole AI Request Workflowhttps://cuespec.ai/blog/compare-typing-and-dictation-workflow/https://cuespec.ai/blog/compare-typing-and-dictation-workflow/Compare typing and dictation with a task-matched worksheet that includes preparation, correction, review and a consistent ready-to-send finish line.Turn a Screenshot and Spoken Feedback into a Clear UI Requesthttps://cuespec.ai/blog/ui-feedback-with-screenshot-and-voice/https://cuespec.ai/blog/ui-feedback-with-screenshot-and-voice/Pair screenshot regions with spoken feedback, a copyable request template and three complete fictional UI change examples.Review an AI Code Change Against the Request You Actually Madehttps://cuespec.ai/blog/review-ai-code-change-against-request/https://cuespec.ai/blog/review-ai-code-change-against-request/Match a requested behavior to a fictional Python patch, example assertions and a review worksheet before accepting an AI code change.Understand an Unfamiliar Codebase Before Asking an Agent to Change Ithttps://cuespec.ai/blog/understand-codebase-before-agent-changes/https://cuespec.ai/blog/understand-codebase-before-agent-changes/Trace one path through an unfamiliar repository, map file evidence and unknowns, and write a bounded follow-up change request.Give a Coding Agent Enough Debugging Context Without Dumping Everythinghttps://cuespec.ai/blog/minimal-useful-debugging-context/https://cuespec.ai/blog/minimal-useful-debugging-context/Select debugging evidence by the question it answers, with three context packets, item-level decisions, and a complete parser investigation request.Turn One Failing Test into a Bounded Agent Requesthttps://cuespec.ai/blog/failing-test-to-bounded-agent-request/https://cuespec.ai/blog/failing-test-to-bounded-agent-request/Connect a failing assertion to its requirement, distinguish implementation and expectation errors, and bound a small repair with meaningful cases.When Local Tests Pass but CI Fails: Give the Agent an Environment Comparisonhttps://cuespec.ai/blog/local-tests-pass-ci-fails-agent-brief/https://cuespec.ai/blog/local-tests-pass-ci-fails-agent-brief/Compare same-revision local and CI conditions, distinguish fixture-path and cache hypotheses, and request one discriminating check.Worktree Isolation and Conversation Context Solve Different Problemshttps://cuespec.ai/blog/worktree-isolation-vs-conversation-context/https://cuespec.ai/blog/worktree-isolation-vs-conversation-context/Check code and conversation mappings separately with a two-task diagram, four mismatch cases, inspection commands and a filled example.Split Parallel Agent Work Around Deliverables and Dependencieshttps://cuespec.ai/blog/parallel-agent-deliverables-and-dependencies/https://cuespec.ai/blog/parallel-agent-deliverables-and-dependencies/Split a feature into four deliverables with a dependency graph, ownership cards, shared fixtures and a corrected parallel-work sequence.Integrate Parallel Agent Changes Without Losing File Ownershiphttps://cuespec.ai/blog/parallel-agent-file-ownership-and-integration/https://cuespec.ai/blog/parallel-agent-file-ownership-and-integration/Combine two agent change proposals using a shared baseline, ownership table, configuration decisions, exact patches and combined checks.Choose a Dictation Route for AI Work on Machttps://cuespec.ai/blog/choose-ai-dictation-input-route/https://cuespec.ai/blog/choose-ai-dictation-input-route/Choose a dictation route for developer prompts on Mac. Compare four named options, destinations, processing stages and when typing is enough.Turn Conflicting Spoken Feedback into One Agent Briefhttps://cuespec.ai/blog/conflicting-feedback-to-agent-brief/https://cuespec.ai/blog/conflicting-feedback-to-agent-brief/Keep three feedback sources separate, record decisions and one unresolved conflict, then build an agent brief from settled constraints.Describe Loading, Cancellation and Late Results Before an Agent Changes the UIhttps://cuespec.ai/blog/loading-cancellation-and-late-results-ui/https://cuespec.ai/blog/loading-cancellation-and-late-results-ui/Specify a six-event UI timeline, result ownership and cancellation behavior before changing how a panel handles late responses.Separate the Feature You Need from the Implementation You Suggesthttps://cuespec.ai/blog/feature-need-vs-implementation-plan/https://cuespec.ai/blog/feature-need-vs-implementation-plan/Separate an export feature's user outcome from technical guesses, then compare two implementation plans against the same acceptance examples.Brief an Agent on a Dependency Upgrade Without Losing Compatibilityhttps://cuespec.ai/blog/dependency-upgrade-compatibility-brief/https://cuespec.ai/blog/dependency-upgrade-compatibility-brief/Map direct and transitive dependency changes, preserve the caller contract, and specify compatibility checks and rollback conditions.Ask an Agent to Repair a Form: Labels, Keyboard Use and Useful Errorshttps://cuespec.ai/blog/accessible-form-repair-agent-brief/https://cuespec.ai/blog/accessible-form-repair-agent-brief/Repair a fictional form with complete before/after HTML, label and error mappings, a copyable agent request and a keyboard walkthrough.Tell a Coding Agent Whether Retry Means the Same Job or Another Onehttps://cuespec.ai/blog/retry-same-action-or-new-one/https://cuespec.ai/blog/retry-same-action-or-new-one/A worked request-identity brief covering lost responses, deliberate second jobs, delayed replays and mismatched inputs.Specify What Undo Restores After an Agent Adds an Actionhttps://cuespec.ai/blog/specify-what-undo-restores/https://cuespec.ai/blog/specify-what-undo-restores/Define Undo for a named archive action with a complete history, preserved later edits, conflict and failure cases, and a copyable agent brief.A CSV Import Brief with Real Edge Cases, Not Just a Happy Pathhttps://cuespec.ai/blog/csv-import-contract-and-edge-cases/https://cuespec.ai/blog/csv-import-contract-and-edge-cases/Define CSV fields, encoding, duplicates and failure policy with nine original fixtures, actual local results and a copyable agent brief.Describe an API Change as a Contract the Agent Can Checkhttps://cuespec.ai/blog/api-change-contract-agent-brief/https://cuespec.ai/blog/api-change-contract-agent-brief/Turn an API change into a checkable contract with request and response examples, compatibility boundaries and an OpenAPI 3.1.1 teaching document.Specify Who Can Read and Change a Recordhttps://cuespec.ai/blog/specify-who-can-read-and-change-a-record/https://cuespec.ai/blog/specify-who-can-read-and-change-a-record/Define record-level read and change permissions with a role matrix, adverse cases, membership withdrawal history and a bounded agent brief.Explain a Partly Successful Batch Before Repeating Ithttps://cuespec.ai/blog/partly-successful-batch-before-repeating/https://cuespec.ai/blog/partly-successful-batch-before-repeating/Use a five-item outcome ledger to separate success, skipped, failed and unknown effects, then define a bounded correction and reconciliation plan.Define a Unique Order Before Asking for Paginationhttps://cuespec.ai/blog/define-unique-order-before-pagination/https://cuespec.ai/blog/define-unique-order-before-pagination/Specify tied-value ordering and retained versus live pagination with eight fictional records, complete pages, data changes, and a copyable agent brief.Keep Units and Rounding Separate in a Coding Briefhttps://cuespec.ai/blog/units-and-rounding-in-coding-brief/https://cuespec.ai/blog/units-and-rounding-in-coding-brief/Define length units, exact input strings, canonical storage and display rounding with nine worked examples and an edit-preservation scenario.Turn File Upload Rules into Observable Outcomeshttps://cuespec.ai/blog/turn-file-upload-rules-into-observable-outcomes/https://cuespec.ai/blog/turn-file-upload-rules-into-observable-outcomes/Define file-upload outcomes with nine fictional files, seven states, inclusive byte limits, expiry and a copyable attachment brief.Specify an Offline Edit Conflict Before Choosing a Mergehttps://cuespec.ai/blog/specify-an-offline-edit-conflict-before-choosing-a-merge/https://cuespec.ai/blog/specify-an-offline-edit-conflict-before-choosing-a-merge/Work through offline edit conflicts with base versions, field changes, lost responses and an explicit reviewed resolution.Scope Keyboard Shortcuts Around Focus and Compositionhttps://cuespec.ai/blog/keyboard-shortcuts-focus-and-composition/https://cuespec.ai/blog/keyboard-shortcuts-focus-and-composition/Define keyboard shortcuts with a focus and composition matrix, twelve key events and a bounded implementation request.Describe a Modal as an Opening-to-Return Focus Flowhttps://cuespec.ai/blog/modal-opening-to-return-focus-flow/https://cuespec.ai/blog/modal-opening-to-return-focus-flow/Specify modal opening, keyboard movement, dismissal and return focus through a complete fictional item-details flow.Decide Whether User Content Is Literal Text or Markuphttps://cuespec.ai/blog/literal-text-or-markup-agent-brief/https://cuespec.ai/blog/literal-text-or-markup-agent-brief/Choose literal text or markup, then define input-to-display boundaries using harmless cases and a plain-text rendering brief.Define a Background Job Beyond Its Acceptance Receipthttps://cuespec.ai/blog/background-job-status-and-result-contract/https://cuespec.ai/blog/background-job-status-and-result-contract/A worked three-stage job contract with changing progress totals, private partial outputs, cancellation races and result retrieval.Apply Incoming Events Without Treating Arrival Order as Truthhttps://cuespec.ai/blog/incoming-events-version-and-reconciliation/https://cuespec.ai/blog/incoming-events-version-and-reconciliation/A complete fictional receiver history for duplicate deliveries, stale snapshots, conflicting event IDs and explicit reconciliation.Give a Rate-Limited Request a Finite Retry Budgethttps://cuespec.ai/blog/finite-retry-budget-for-rate-limited-reads/https://cuespec.ai/blog/finite-retry-budget-for-rate-limited-reads/Work through seconds and HTTP-date Retry-After schedules with request time, wait time, attempt limits and a clear stop state.An AI Localization Brief That Preserves Keys, Placeholders and Pluralshttps://cuespec.ai/blog/localization-brief-keys-placeholders-plurals/https://cuespec.ai/blog/localization-brief-keys-placeholders-plurals/Preserve message keys, arguments and plural behavior when briefing an AI agent on localization, with English and Chinese catalogs and a complete brief.Explain Time-Zone Requirements Before Asking an Agent to Fix Dateshttps://cuespec.ai/blog/timezone-requirements-before-date-fixes/https://cuespec.ai/blog/timezone-requirements-before-date-fixes/Distinguish instants, local wall times and dates before asking an agent to fix time-zone behavior, with daylight-saving examples and a complete brief.A Feature-Flag Brief That Defines Defaults, Failures and Rollbackhttps://cuespec.ai/blog/feature-flag-defaults-failures-rollback/https://cuespec.ai/blog/feature-flag-defaults-failures-rollback/Specify flag defaults, readiness, errors and rollback before implementation, with a four-state example and a manual rollout decision template.Keep Developer Documentation Aligned with an Agent's Code Changehttps://cuespec.ai/blog/documentation-after-agent-code-changes/https://cuespec.ai/blog/documentation-after-agent-code-changes/Map an agent's configuration change to a complete README update, with a worked source example, migration note, and honest command status.Record an Architecture Decision So the Next Agent Knows Whyhttps://cuespec.ai/blog/architecture-decisions-for-the-next-agent/https://cuespec.ai/blog/architecture-decisions-for-the-next-agent/Write an architecture decision record with alternatives and consequences, then preserve its history when a later decision supersedes it.Ask an Agent to Investigate Slow Code Before Requesting a Rewritehttps://cuespec.ai/blog/slow-code-investigation-before-rewrite/https://cuespec.ai/blog/slow-code-investigation-before-rewrite/Investigate slow Python code with a fixed workload, hypothesis tree, versioned profiling commands and an empty baseline/candidate record.Write a Pull-Request Description That Links the Change to Evidencehttps://cuespec.ai/blog/pull-request-description-with-evidence/https://cuespec.ai/blog/pull-request-description-with-evidence/Turn a small patch into a pull-request description that distinguishes code changes, planned checks, and observed verification.Continue the Agent Conversation or Start a New Task?https://cuespec.ai/blog/continue-agent-chat-or-start-new-task/https://cuespec.ai/blog/continue-agent-chat-or-start-new-task/Use three worked scenarios and ready-to-copy requests to decide whether to continue an agent conversation, start a new task, or clarify its state.Brief a Stored-Field Migration Across Old and New Readershttps://cuespec.ai/blog/stored-field-migration-old-new-readers/https://cuespec.ai/blog/stored-field-migration-old-new-readers/Before asking an agent to rename a stored field, describe the readers and writers that can still use its old name. Then define how existing records move, how conflicting values are resolved, and when returning to the old application stops being a simple code rollback.Specify Redirects, Missing Routes and Allowed Methodshttps://cuespec.ai/blog/redirects-missing-routes-allowed-methods/https://cuespec.ai/blog/redirects-missing-routes-allowed-methods/Before asking an agent to “fix the old URLs,” list the requests those URLs receive. A moved reading page, a POST endpoint and an unknown path need different outcomes. Give each request a status, a destination when appropriate, and a rule for its method, body and query parameters.Choose Search Matching Rules Before Adding Fuzzinesshttps://cuespec.ai/blog/unicode-search-matching-before-fuzziness/https://cuespec.ai/blog/unicode-search-matching-before-fuzziness/Before asking an agent to make search “more forgiving,” decide which differences should count as equal. Two labels can look the same while using different Unicode sequences. Other labels differ in case, accents or character width. Treating all those differences as one problem makes the search requirement difficult to review.Keep a Long Spoken Coding Request Recoverable Before You Send Ithttps://cuespec.ai/blog/keep-long-spoken-coding-request-recoverable/https://cuespec.ai/blog/keep-long-spoken-coding-request-recoverable/Keep reviewed parts of a spoken request as separate text notes before assembling the message you will send. Record an explicit correction beside the earlier clause it replaces. If a part has no retained wording, leave a visible gap and ask for a new decision instead of producing a smooth but invented request.Switch a Technical Task to Voice Without Guessing What Carries Overhttps://cuespec.ai/blog/switch-technical-task-to-voice-continuity/https://cuespec.ai/blog/switch-technical-task-to-voice-continuity/When you want to speak instead of type during a technical task, first decide where the answer should come from. You can use speech to prepare text for an existing task, or use a conversational voice route that receives speech and responds. Those choices can look similar while leaving different things visible.