Guide / Updated July 2026
20 AI Tricks You Can Use Today
The evergreen tactics that still pull their weight, plus nine new tricks — dreaming agents, outcome rubrics, dynamic workflows, and more.
Prompt library / Copy, adapt, run
A better starting point than a blank box.
Twenty complete prompts for connected AI tools. Replace the words in brackets, keep the guardrails, and make the instruction yours.
Inspired by OpenAI's prompt boardTurn everything you missed into one calm catch-up.
Review my Slack, Gmail, and Google Calendar activity since [time or date]. Give me three short sections: Act now, Needs a decision, and Good to know. For every item, explain what changed and link to the original source. Ignore repeated or unchanged threads. Do not reply, schedule, or update anything without my approval.
What makes this useful
- It defines a precise time window instead of asking for a vague summary.
- The three output buckets turn information into an immediate decision queue.
- Source links make every claim easy to verify before you act.
Try next
Turn that catch-up into a plan for the rest of today. Put no more than three items in Must do, then list what can safely wait and why.
Reset your inbox without losing control of it.
Review my Gmail inbox and sent mail from the past [number] days. Identify messages that genuinely need my attention, group the rest into sensible cleanup categories, and draft replies in my voice using relevant context from Slack, Google Calendar, and Google Drive. Show me the proposed replies and cleanup plan first. Ask before sending, archiving, labelling, or moving anything to Trash.
What makes this useful
- Sent mail gives the model evidence of your actual tone and commitments.
- Triage and drafting happen together, so important context is not separated.
- The approval boundary keeps irreversible inbox actions in your hands.
Give one project a thread that remembers what matters.
Treat this conversation as the working thread for [project or launch]. Review the relevant Slack discussions, Google Drive documents, Gmail messages, and Google Calendar events. Report meaningful changes, blockers, commitments at risk, missing owners, and decisions that are overdue. Link each point to its source and separate confirmed facts from assumptions. Check again every weekday at [time], but only notify me when something materially changes. Never message people or edit project records without approval.
What makes this useful
- A dedicated thread preserves project context instead of rebuilding it each time.
- Material-change filtering prevents routine updates from becoming more noise.
- Facts and assumptions are separated before they reach a decision-maker.
Write the weekly update from the work you actually did.
Use Slack, Gmail, Google Calendar, and Google Drive to reconstruct my past seven days. Draft a concise weekly update covering what shipped, progress made, decisions taken, blockers, and next steps. Match the tone and structure of my previous updates, include links to the strongest source for each claim, and flag anything you could not verify. Do not post or send the update without my approval.
What makes this useful
- The draft is grounded in activity rather than an unreliable end-of-week memory.
- Previous updates act as a real style sample, not a generic tone instruction.
- Unverified claims are surfaced before they become accidental status theatre.
Try next
Rewrite this as a five-line update for my manager: one outcome, two progress points, one blocker, and one next step. Keep every source link.
Turn repeated feedback into a playbook you can reuse.
Review the feedback I have received in Slack and comments on documents in Google Drive. Find recurring preferences, standards, corrections, and examples of what was praised. Turn the stable patterns into a short reusable playbook for future writing, presentations, and project updates. Include before-and-after examples where the evidence supports them. Preserve my voice, cite the source material, and list any contradictory guidance that needs clarification.
What makes this useful
- Repeated feedback becomes an asset instead of disappearing into old threads.
- Examples make abstract preferences usable on the next piece of work.
- Contradictions stay visible rather than being averaged into bad advice.
Resolve the clear document comments and isolate the hard ones.
Open [Google Doc link or title] in Google Drive and review every unresolved comment. Apply edits only when the requested change is clear and consistent with the document’s voice and structure. Leave conflicting, ambiguous, or high-impact comments open. Then give me a change log with three groups: edited, needs a decision, and could not be completed. If live editing is unavailable, show the proposed edits instead.
What makes this useful
- Clear edits move forward while judgment calls remain visible to a human.
- A change log makes the work auditable without rereading the full document.
- The fallback still produces value when direct editing is unavailable.
Turn your best deck or workbook into the next team template.
Use Template Creator to convert [approved PowerPoint deck or Excel workbook] into a reusable team template. Preserve the structure, layouts, formulas, validation rules, and visual language that should remain consistent. Replace client- or project-specific material with clear placeholders and guidance. Then create a fresh version using [brief, report, or source data]. Summarise what was preserved, what changed, and what still needs review.
What makes this useful
- It distinguishes the reusable system from one project’s accidental details.
- Placeholders carry the original logic into the next piece of work.
- The test run proves the template works before the team depends on it.
Make a messy spreadsheet trustworthy before analysing it.
Review [spreadsheet, CSV, or Drive file] without overwriting the original. Create a clean copy that standardises dates and categories, identifies duplicates, and handles missing or invalid values transparently. Produce a short data-quality log, then build a small set of clear charts for the most decision-relevant trends or surprises. Finish with the main takeaway, the limits of the data, and the items I should verify manually.
What makes this useful
- Cleaning is documented separately from analysis, protecting the original evidence.
- Charts are tied to decisions rather than generated for decoration.
- Known limits and manual checks stop a polished output from overstating certainty.
Prepare a time-off request across the tools your team uses.
Use Computer with my signed-in browser to prepare a time-off request in [HR system] for [dates]. Check Google Calendar for conflicts, draft an out-of-office calendar hold, and prepare a short Slack note for my team with any coverage details. Show me every proposed change and any conflict you found. Stop before submitting the request, creating the event, or sending the message.
What makes this useful
- One instruction coordinates the request, calendar, and team communication.
- Conflict checking happens before the absence is announced.
- Every external action has an explicit stop point for review.
Turn a receipt into an expense report ready for review.
Find the receipt for [purchase] in Gmail, locate the relevant expense policy in Google Drive, and use Computer with my signed-in browser to prepare the claim in [expense system]. Cross-check the date, amount, currency, category, tax, and required evidence. Show me the receipt, every completed field, anything missing, and any policy risk. Do not submit the expense without my approval.
What makes this useful
- The receipt and policy are checked before fields are entered.
- Missing evidence is surfaced while it is still easy to fix.
- Preparation is automated while submission stays a human decision.
Watch for a genuinely good cinema seat—not just any seat.
Use Browser to check the live seat maps for [film] at [cinema] on [date range]. Look only for [preferred section or rows], exclude accessible and companion seats unless I explicitly request them, and check Google Calendar before recommending a showtime. If no suitable seat is available, check again every [interval] and notify me only when a new match appears. Include the showtime, seat position, total price, and booking link. Never purchase or change a booking without asking.
What makes this useful
- A subjective idea of a good seat becomes a testable search rule.
- Calendar context removes options that are available but unusable.
- The monitor stays quiet until the result actually changes.
Keep watch for the sold-out thing in the right size and price.
Use Browser to find [item] in [size, colour, or specification] for no more than [maximum total price]. Check the retailer and credible resale listings, confirming actual availability, condition, returns, shipping, and the final delivered price. If nothing qualifies, check once a day and alert me only when a real match appears. Include the direct listing link and the trade-offs. Do not buy anything or contact a seller without permission.
What makes this useful
- The search uses total cost and real availability rather than headline price.
- Condition and return terms make resale options comparable to retail.
- A scheduled check replaces repetitive manual browsing.
Find homes that match your life, not merely your filters.
Use Browser to find current homes matching [location, budget, move-in date, and non-negotiables]. Open the original listings and compare total monthly cost, availability, layout, commute to [place], and likely deal-breakers. Give me a ranked shortlist with direct links and explain the trade-off behind each rank. Check again every morning and notify me only when a stronger match appears. Do not contact agents, book viewings, or submit an application without approval.
What makes this useful
- The shortlist includes lived experience factors, not just listing filters.
- Rankings expose the trade-offs instead of pretending one option is perfect.
- The daily monitor reports improvements rather than repeating the same inventory.
Find a better flight and keep watching the real total cost.
Use Gmail to find the details of my trip, Google Calendar to confirm when I need to arrive, and Browser to compare current flights for [route or dates]. Show the total price including baggage, airports, door-to-door travel time, layovers, change terms, and the most important inconvenience. Check each morning for a meaningfully better price or schedule and tell me what improved. Do not book, cancel, or change any reservation without my approval.
What makes this useful
- Arrival needs and existing bookings are recovered before comparison begins.
- Total journey cost prevents a cheap fare from hiding expensive friction.
- Future alerts explain the improvement instead of reporting every fluctuation.
Make group accommodation options easy to compare honestly.
Use Browser to find live accommodation options for [destination, dates, group size, and budget]. Open each original listing and verify the total price, availability, real sleeping arrangements, bathrooms, accessibility needs, location, travel time, cancellation policy, and house rules. Create a concise comparison with listing links and photos, then explain which options genuinely fit the group and where each compromise sits. Do not book or contact a host without approval.
What makes this useful
- Sleeping arrangements and accessibility are verified instead of inferred from capacity.
- A shared comparison replaces an unstructured pile of links.
- Trade-offs are explicit before a group starts voting.
Find the photographs worth keeping without touching the originals.
Use Google Drive to review [photo folder], including available dates, locations, and camera metadata. Select the strongest images, group them by place or moment, and separately flag blur, screenshots, duplicates, and near-identical frames. Create a contact sheet or illustrated shortlist that links back to each original. Explain the selection criteria. Do not delete, move, rename, edit, or share any file without asking first.
What makes this useful
- Selection and cleanup candidates are kept as two separate judgments.
- Metadata helps recover the story of a large, messy folder.
- Original files remain untouched while the shortlist stays easy to navigate.
Plan the week around clothes you already own.
Use my wardrobe photos in Google Drive, relevant purchase details in Gmail, my Google Calendar for the next seven days, and Browser for the weather. Suggest an outfit for each day using only items you can identify in my wardrobe. Explain why it fits the event and conditions, offer one backup where useful, and ask about anything uncertain. Do not buy anything or share my photos without permission.
What makes this useful
- Recommendations are grounded in your real wardrobe and actual week.
- Weather and event context turn styling into a practical plan.
- Unidentified items trigger a question instead of an invented assumption.
Check whether your finances need attention without oversharing them.
Use Finances to prepare a private, high-level check-in. Flag upcoming bills and renewals, recurring subscriptions, unusual category changes, and possible cash-flow pressure. Compare recent spending patterns with my normal baseline and suggest a few practical questions or savings opportunities. Do not display account numbers, exact balances, merchant names, or individual transactions. Never move money, cancel a service, or take any external action without asking me first.
What makes this useful
- The output focuses on decisions while minimising sensitive detail.
- Changes are compared with your own baseline rather than a generic budget.
- The prompt draws a hard line between insight and financial action.
Build a coach that reads the training log before prescribing.
Act as my long-term training coach and journal manager. First interview me in small groups of questions about goals, history, schedule, equipment, preferences, sports, injuries, and current symptoms. Then create a weekly structure and an Excel workbook as the permanent source of truth. Before every recommendation, read and update that workbook with my completed work, readiness, soreness, symptoms, location, and lessons. Prescribe only the next workout and explain its role in the larger plan. Never claim something is logged unless the workbook was updated, and stop for medical review when symptoms make training unsafe.
What makes this useful
- The workbook prevents coaching from depending on fragile conversational memory.
- One-workout-at-a-time guidance can adapt to readiness and real life.
- Logging rules and safety limits reduce confident but context-free advice.
Wake up to a small newspaper made for your actual interests.
Every morning at [time], use Browser to create a concise personal briefing about [cities, teams, artists, hobbies, markets, local events, and other interests]. Include only developments that are genuinely new or useful, explain why each one matters to me, and link to the strongest original source. Avoid repeated stories, stale updates, and low-signal commentary. If Google Calendar is connected, add anything that could affect today’s plans. Format it for quick reading on a phone.
What makes this useful
- A personal relevance test keeps the briefing smaller than a normal news feed.
- Original links make it easy to go deeper without bloating the summary.
- Calendar context connects information to what you are actually doing today.
20 AI Tricks You Can Use Today
Refreshed for July 2026. The evergreen tactics that still pull their weight, plus nine new tricks that simply weren't possible in March — dreaming agents, outcome rubrics, dynamic workflows, and more.
Each trick includes the exact prompt to use and how to set it up. The "New in July" tag marks what shipped since the March edition. Organized into seven categories: Productivity, Mobile & Remote, Self-Improving Agents, Orchestration at Scale, Content & Design, Prompting & Quality, and Models & Business.
Productivity & Everyday Automation
Inbox Sweep: Triage 100+ Emails in One Command
Instead of reading every email yourself, tell Claude to scan your entire inbox, identify which emails actually need your reply, draft responses for each one, and generate a dashboard showing everything at a glance. What used to take 30–45 minutes becomes a single command.
How to set this up
- Connect Gmail to Claude via Connectors (Settings → Connectors → Gmail)
- Open Cowork and select your working folder
- Paste the prompt above — or just say "inbox sweep" if you have the skill installed
- Review the HTML dashboard that appears in your folder
Meeting Prep Briefing in 60 Seconds
Walking into a meeting unprepared? Claude can pull your calendar, check your email history with each attendee, find relevant documents, and produce a one-page briefing with background on who you're meeting, what you discussed last time, your open action items, and suggested talking points.
How to set this up
- Connect Google Calendar and Gmail via Connectors
- Optionally connect Google Drive for document access
- Run the prompt before your next meeting (or trigger via Dispatch from your phone on the way there)
Receipt Scanner to Instant Expense Report
Drop a folder of scanned PDF receipts onto your desktop. From your phone, tell Claude to turn them into a categorized expense report. It reads every receipt, extracts amounts and vendors, categorizes spending, and generates an interactive HTML report with charts and breakdowns.
How to set this up
- Create a receipts folder in your Cowork project
- Scan or save your receipts as PDFs into that folder
- Run the prompt from Cowork or Dispatch
Replace Notion with a Folder + Claude
Create an empty folder. Tell Claude to set itself up as your personal knowledge assistant with an orchestrator agent, a local SQLite database for structured data, and a simple HTML interface to browse it. You now own 100% of your data, it works offline, and you can swap AI models anytime.
How to set this up
- Create a new empty folder on your desktop (e.g., "PKA")
- Open it in Claude Code (cd ~/Desktop/PKA && claude) or select it in Cowork
- Paste the prompt and let Claude set up the entire structure
- Drop files into the Team Inbox — Claude auto-organizes them
Mobile & Remote AI
Run 5 Tasks in Parallel from Your Phone
Claude Dispatch lets you fire off multiple independent tasks from your phone. Each runs as a separate parallel agent on your desktop. While you're at the gym, one agent sweeps your inbox, another researches a topic, a third preps your meeting, and a fourth generates a presentation.
How to set this up
- Set up Dispatch (Cowork → Dispatch → scan QR code with phone)
- Keep your computer awake (System Settings → Energy → Prevent sleeping)
- Send the prompt from the Claude mobile app
- Each task runs independently — results appear on both phone and desktop
Connect Claude to 8,000+ Apps via Zapier MCP
Claude's native connectors cover about 38 apps. But using Zapier's MCP server, you can extend that to 8,000+ apps with 30,000+ actions. Connect to School, Airtable, HubSpot, Stripe, or any app Zapier supports — then trigger actions from Claude or even from your phone via Dispatch.
How to set this up
- Go to Zapier → MCP → Start Building → New MCP Server
- Select "Claude" as your client
- Search and add tools for any apps you want (e.g., School, HubSpot, Airtable)
- Copy the generated URL
- In Claude: Connectors → Zapier → paste the URL
- Select "Always allow" for the tools you added
Self-Improving Agents — New for 2026
Let Your Agents Dream — and Write Their Own Playbooks
Dreaming is the biggest shift since agents themselves. Between sessions, an idle agent reviews everything it did in its last jobs, pulls out the patterns — what worked, what kept failing — and writes new memory entries and playbook files the next session can use. Anthropic compares it to how your brain consolidates memory during sleep. Legal AI company Harvey reported task completion rates up roughly 6x after enabling it. Your agents stop making the same mistake twice — without you writing a single instruction.
How to set this up
- On Claude Managed Agents: enable Dreaming in your agent settings — it runs on a schedule while agents are idle
- Review the generated playbook .md files weekly and prune anything stale
- Solo users: use the DIY prompt above — same principle, manual trigger
Grade the Work, Not the Prompt: Outcomes
Stop writing longer prompts and start writing rubrics. With Outcomes, you define what a successful result looks like, and a separate grader — running in its own context window, uninfluenced by the agent's reasoning — evaluates the output against your criteria. If something falls short, the grader pinpoints what, and the agent takes another pass. Wisedocs cut document review time by 50% with this. It's the single best answer to "the output is almost right but not quite."
How to set this up
- On Managed Agents: define an Outcome rubric in the agent config — the independent grader loop is built in
- Everywhere else: use the DIY prompt — self-grading against an explicit rubric catches most of the gap
- Keep rubrics to 3–5 criteria. More than that and the grader's signal gets mushy
Put Your Agents on a Schedule with Routines
Agents no longer need you to press the button. Routines let you schedule recurring agent work: a 7am daily briefing that's waiting when you wake up, a Friday afternoon weekly report, a Monday morning pipeline review. Combined with Dreaming, a routine agent gets better at its recurring job every single week — it's the compounding loop I keep writing about, running on autopilot.
How to set this up
- In Cowork: set up a Routine with a schedule and a standing prompt
- In Claude Code: cron-style scheduled sessions do the same job
- Start with one routine you'd otherwise do manually every day — the morning briefing is the classic
Orchestration at Scale — New for 2026
Dynamic Workflows: Hundreds of Agents, One Command
This wasn't possible in March. Claude Code can now write its own orchestration script — breaking a big job into subtasks, fanning out tens to hundreds of parallel subagents, and adversarially verifying results before anything reaches you. The proof point: Bun's creator used dynamic workflows to port the entire runtime from Zig to Rust — roughly 750,000 lines, 99.8% of the test suite passing, eleven days from first commit to merge. For your world: full-codebase audits, mass content reviews, bulk data migrations.
How to set this up
- Update Claude Code and just describe the fan-out you want — "use a workflow" opts you in
- Best fits: jobs that decompose into many similar pieces (per-file, per-module, per-record)
- Always ask for a verification stage — parallel agents are fast, adversarial checking is what makes them trustworthy
Hierarchical Agent Teams, Three Levels Deep
Claude Code agents can now spawn child agents up to three levels deep: a lead agent delegates to per-area leads, which delegate to per-file workers. Each specialist gets its own context, prompts, and tools; results flow back up the tree. The practical difference from flat parallelism: the middle layer can coordinate, dedupe, and quality-check its workers before anything hits the lead — like a real org chart, minus the meetings.
How to set this up
- Describe the org structure in your prompt — lead, coordinators, workers
- Give the middle layer an explicit review job, not just routing
- Watch the audit trail to see what each sub-agent did, in what order
Turn a Working Session into a Live, Shareable Dashboard
Claude Code sessions can now publish their output as a live Artifact page on claude.ai — one that updates in place as the session keeps working. Kick off a long-running analysis, share the link with your team, and they watch the report fill in as the agents progress. No more "I'll send you the results when it's done" — the results are a URL from minute one.
How to set this up
- Currently in beta on Team and Enterprise plans
- Ask the session to create an artifact and keep updating the same page
- Artifacts are private by default — you choose when to share the link
Content, Design & Brand
Clone Your Writing Voice from AI Memory
Claude accumulates memory of your conversations over time. Ask it to create tone-of-voice guidelines based on what it's learned about how you communicate. This becomes a reusable markdown file that every AI skill references, ensuring all content sounds like you.
How to set this up
- Have several conversations with Claude first so it learns your style
- Run the prompt above to generate the guidelines
- Review and refine the output
- Save it as a skill markdown file so all future content references it
On-Brand Design Without a Designer: Claude Design
Claude Design now sticks to your design system across projects: feed it your colors, type, and components once, and every asset it generates stays on-brand. You can edit directly on the canvas instead of re-prompting, and it works fluidly with Claude Code — design a component visually, then have Code wire it into your actual site. For the 80% of design work that doesn't need a professional, this replaces the entire brief-wait-revise cycle.
How to set this up
- Find Claude Design in the sidebar of the Claude desktop app
- Upload your brand kit or design tokens once — it persists across projects
- Edit results directly on the canvas instead of re-prompting from scratch
- Hand finished designs to Claude Code to implement on your site
Prompting & AI Quality Control
Force Blank Answers to Kill Hallucinations
When AI extracts data from documents, it guesses instead of admitting it doesn't know. Fix this by explicitly giving AI permission to leave fields blank and requiring an explanation for each blank. You only review blanks instead of everything.
The "3x Penalty" Trick: One Line That Changes Everything
AI defaults to guessing because it treats a wrong answer the same as a blank answer. Add a single line that changes the incentive: tell AI that a wrong answer costs 3x more than saying "I don't know." This dramatically reduces hallucinations. Like telling a new hire: "If you give me wrong info it costs the company 3x more than just saying you'll check."
That's it. One line. Add it to any extraction, analysis, or review prompt. Combine with Trick #15 for maximum effect.
Source Tagging: Catch AI When It Infers Instead of Extracts
Even after you tell AI to only extract from the document, it will start inferring on complex tasks. This safety net catches it. Require a "Source" column on every field with two possible values: Extracted (word-for-word from the document) or Inferred (derived from context). For inferred values, require a one-sentence evidence explanation. Now you only need to review the inferred fields.
The complete anti-hallucination stack: Combine Tricks #15 + #16 + #17 in every data extraction prompt. You'll go from checking everything to only checking blanks and inferred fields.
Plan Mode: Make AI Explain Before It Acts
Before Claude does anything complex, switch to Plan Mode. Instead of executing immediately, it asks clarifying questions, shows you a step-by-step plan, and waits for your approval. This prevents expensive mistakes, runaway agents, and "overkill" implementations. In Claude Code, press Shift+Tab to toggle. In Cowork, add "plan first, don't execute yet" to your prompt.
Models & Business
Right-Size Your Model: Sonnet 5 by Default, Fable 5 for the Hard Stuff
Two model moves worth acting on. First: Sonnet 5 launched in June as the most agentic model yet, with introductory pricing ($2/$10 per million tokens) through August 31 — go re-run the automations that failed for you in February; several will just work now. Second: Fable 5, the new tier above Opus, went globally available July 1. The trick is routing, not loyalty: cheap fast models for mechanical work, Sonnet 5 as your default, and Fable 5 reserved for the genuinely hard reasoning — the gnarly contract, the architecture decision, the analysis you'd otherwise hire for.
How to set this up
- Keep a "failed prompts" folder — re-run it on every major model release
- Default to Sonnet 5 for daily agent work (it's the Free/Pro default now anyway)
- Escalate to Fable 5 only when the task is worth the premium — hard reasoning, high stakes
- Review your routing quarterly: yesterday's premium task is today's default-tier task
Claude Finance: 10 Pre-Built Agents for Your Back Office
Anthropic shipped Claude Finance — a package of ten predefined agents for financial workflows: a pitch builder, meeting preparer, market researcher, evaluation reviewer, and month-end closer, among others. Built for financial services, but here's the SMB trick: the month-end closer and market researcher work just as well for a normal company's books and competitive landscape. You're getting workflows that a consultant would charge five figures to design, pre-built.
How to set this up
- Available as part of Claude's managed agent offerings — start with one agent, not all ten
- The meeting preparer and market researcher generalize far beyond finance
- Feed each agent one past example of the output you want (last month's close, last quarter's research)