claudeers.
// Claude Skills

awesome-prompts

434 awesome one-line power prompts for coding agents (Claude Code, Codex, Cursor) - density beats length

Install with your AI

Paste into Claude Code, Cursor, or any agent — it reads the repo and wires the tool into your project.

Install and set up awesome-prompts (claude-skill project) into my current project.
Found on https://claudeers.com/awesome-prompts-2
Repo: https://github.com/Drix10/awesome-prompts
Homepage/docs: https://github.com/Drix10/god-tier-mini-prompts/blob/main/SKILL.md
Detected install method: claude-skill → # copy this skill into .claude/skills/awesome-prompts/
Category: skills. Platforms: cli, api, web.
Read the repo's README for exact setup and env vars, then install it and wire it into my project.

Claudeers Health Verdict:
unknown; community-verified: false. Confirm the source before running anything.
// or install directly (claude-skill)
# copy the skill dir into your project:
# .claude/skills/awesome-prompts/   (or ~/.claude/skills/awesome-prompts/ for all projects)
// or clone
git clone https://github.com/Drix10/awesome-prompts

// compatibility

Platformscli, api, web
Operating systems—
AI compatibilityclaude
LicenseMIT
Pricingopen-source
Language—

Get your FREE $2.50 API credits to access TickAtlas financial data ↗

⚡ awesome-prompts

434 one-line power prompts for coding agents. Each sentence so dense every word does work.

Density beats length. One loaded sentence outperforms a page of instructions.


📦 Install

Copy the whole folder - SKILL.md plus references/ - into your agent's skills directory. That's the whole install.

Claude Code: ~/.claude/skills/awesome-prompts/

git clone --depth 1 https://github.com/Drix10/awesome-prompts.git ~/.claude/skills/awesome-prompts

OpenAI Codex: ~/.codex/skills/awesome-prompts/

git clone --depth 1 https://github.com/Drix10/awesome-prompts.git ~/.codex/skills/awesome-prompts

Cursor: reference it from .cursor/rules/ or paste prompts straight into Agent mode.

Then invoke by matching a task to a prompt, or run the meta-prompt and let the agent pick.


Why this exists

A single sentence of vectors so juicy each verb, noun, and adjective acts like a flavor bomb. A simple prompt that ends up being VERY impactful. Short prompts also get a fresh context window's full attention, so delegate the real hunting to subagents and keep the main thread narrow.

This repo is a living collection of those sentences: harvested reply-by-reply from the state-machine thread (136 replies, 193 quotes, plus the replies on the quotes that invited them), the threads it spawned, Prompt Snacks, Notchbird's Banger Prompts, Anthropic's official patterns, and the best engineering threads on X.

The skill stays lean on purpose: SKILL.md is a 50-line router (meta-prompt + index), and the 434 prompts live in twelve references/ files that the agent opens only when the task matches. Credits live in this README, never in agent context. Progressive disclosure, per the skill authoring best practices.

🗂 The collection

Twelve parts, each a standalone file the agent reads on demand. Match the task, open one file, use what fits.

PartCountWhat it steers
🧠 Meta3 + ruleLet the agent pick its own prompts
🏛 Architecture22Abstraction-hunting, system framing
🔍 State-machine audits63The complete implicit-state-machine playbook
🦄 Unicorn hunts19DISCOVER hunts: invariants, transactions, trust
🐛 Debugging66Symptoms vs roots, failure modes, hunts
📐 Planning48Specs, grilling, handoffs, autonomy
✨ Quality & repair67Smaller diffs, TDD, repair sets, security sweep
✅ Verification35Evidence, holdouts, PROVE set, ship decisions
👀 Review23Two-axis reviews, PR shape
🧹 Hygiene3Transcripts, AGENTS.md, usage audits
📚 Learning15Mental models, diagrams, ELI5
🎛 Steering70Morale, BOSS coda, updates, skills
📝 NotesprinciplesField principles - read alongside any part

Total: 434 prompts, plus 3 paste-ready expanded audits and one full phased-audit text.

🪄 The meta-prompt

Don't browse, delegate. Paste this with your task, idea, or code:

Consider all the mini-prompts in this skill, pick the ones that would
improve my idea/input/software the most, and apply them.

Narrow it per phase: "…pick the 3 best for planning / review / debugging this."

🙏 Sources & credits

Every prompt in the skill came from a human. Full credit table:

The state-machine thread (@steveruizok, Oct 2026; seed: "find all of the implicit state machines in my app")

AuthorContribution
@trevinThe expanded one-file audit (states, invariants, transition tables, retries + concurrency, prioritize, group, fix, keep updated)
@NikzaaaaadThe 9 follow-ups (false screens, UI business logic, drifted copies, unenforced DB rules, races, swallowed errors, magic numbers, loose types, effects-as-events)
@flowforth_76The full phased audit (discovery → verification → plan → implement)
@carpenterdroidThe refactor-directive mega-prompt (eliminate state, derive values, explicit machines, regression + integration checks)
@asmirknTransition tables; impossible-but-reachable transitions
@aloobhujiyanWhere two machines touch the same data (the race lives there); model-checker shortest-path
@Outgathr"Which of these explain the bugs I've been calling flaky"
@liuzhao_666Concrete event sequences; permitted vs reachable states
@eclarkhalidCite handlers + conditions; mark proposed changes separately
@akashc777Hiding spots: flags the UI never designed for
@jj_kovacs16 states on paper, 5 real; bugs in the other 11
@kabelsalat_infoList combos that can't happen but code allows
@TalljackcvCancelled must never return to success
@BongJupite43260Cross-component inconsistent paths
@sathya_parthaScreen rules no component enforces
@gippp69Harness rules: allow/ask/deny, BLOCKED, receipts
@xsmryzeEach model only on its best job
@jasonmelloPlan in 5 bullets before anything
Amin RoostaCall a spade a spade
John RushFresh chat after loops
@Voxyz_aiRate bugs 1-10; no designing for rare cases
@_____CalumNarrow beats smart; hints make it worse
@franzejrTickets need behavior, constraints, checks
@MisterSnowCrashRegression test failing-then-passing artifact
@BlossomAILabsReviewer gets acceptance criteria first
@huesofbanterTest for removing workarounds
@cnwenfCheckable handoff results
@itsvdruzkoTwo agents disagree → edit ticket
@z1njoInteraction, not a picture
@choblin29Requests vs hard boundaries
@kr4chininVisualize the state flow
@sjmurdochVerify correctness in Quint
@garbnzghSubagents + propose building it into reviews/workflows
@badbirbFrame out stateful elements and plan before coding
@EternitiesAIFailing regression test = boundary of suspected vs fixed
@Pps831Global-rules prevention; no-implicit-state-machines rule
@feddiwhipUnenforced invariants
@bonsaielectricImplicit invariants
@peytoncasperNon-terminal states variant
@jussisaurThe explicit inversion (turso)
@twid"Find all tautologies"
@wahome_thukuThe legal port: implicit obligations in contracts
@jimlux"Find the dumb requirements" (MCAS debate)
@cocobiskitAdversarial red-teaming
@joshest3p"Pushing determinism down"
@krimuthuAutonomous fix loop ("keep me out of the loop")
@_benwick"Just find the bugs"
@orcsrise"What pattern allowed this bug to exist"
@udaysy"What breaks if we revert this in 6 months"
@ShannoncodeRoot-cause tokens; SDK/backend/schema context assembly
@menhguinScientific method; decision frameworks (Chesterton, OODA, autonomy levels)
@matt_dot_solTargeted "are you sure" double-checks
@PreciousBa82157"If it's not in the prompt, it doesn't exist"
@bigdatachadsResearch-first, then root-cause
@asavs_Roleplay user / agent / system internals
@adityavg13Internalize metaphors (allegory of the cave)
@impervium888Explain it as a story
@codeby_ABIRecursive prompt booster
@0xRajatOne-liners over elaborate skills
@ZazenCodesDoing things wrong vs doing the wrong things
@teodorioMaster-agent routing; vector-path steering
@PawelPSzczesnyMine operators, combine into new prompts
@BadGatewayCoThe combo: machines + invariants + deepen modules
@webtkdevConvert to actors; rules-file ban
@mrcommando712Map states before fixing; late-response check
@PaavoKarlinReport shape: found / decide / bugs / sound
@imahmoodInterview-me + Simplified Technical English
@cryptoquick / @GarrettPetersenExplicit, tested, documented
@feddiwhipThe 16 unicorn hunts + PROVE/REPAIR sets + evidence prompt
@amronos / sprocketCleanup sweep: over-engineering, dead text, duplication, coverageless tests
@GKev1nRead-only guardrail: states, transitions, conflicts + line numbers; wild boolean machines
@buchmansterSpecs first, then tests from specs
@oderajosephScope the first refactor
@Nikzaaaaad9-more list: false screens, UI logic, DB rules, magic numbers, loose types
@VincentLinwxRequest IDs + cancellation vs state
@maxxrubin_Blast-radius thinking per change
@psychic_jammerBan comments except the exceptional
@Lion_GameDevUnmentioned edge cases get assumed
@charleshxpeStructured scan: states, moves, conflicts, risk rank
@0xUltraInstinctMake it permanent in AGENTS.md
@literallyhimmmmRead the bugs first; half aren't real
@alexlee20122022Walk the unhappy paths principle
@kr4chininLeads-not-findings; Confirmed/Likely/Smell; recreate-not-reset
@KelvinW_infoOne named principle per run
@deepman_1Long-running things become explicit machines
@ddyodsRules in the harness, practices in the prompt
@jbaruchRight 300 tokens beat 100k noisy ones
@yanggao97566Machines disagreeing about done
@boringdev77Flags or status strings?
@CarbonMatchAIVocabulary is a senior skill
@IamYashKapoorBooleans secretly meaning step 3
@arpittalksstuffisReallyLoading and isDoneForReal
@asmirknEnum + transition table + failing test
@fishjarUnnamed state bits
@PowerduckLabsRetries plus flags; name and reject
@jeffbruchadoHandler order is the only spec
@OdysseyStackThe unnamed isLoading conjunction
@johnroodepicA type that rejects the fourth flag
@looking5452Model with typestate
@hboonStudy-and-find phrasing; 16 short prompts (elaborate simply, root cause, supply-list, essence, mock, review-context)
@illulachy_Enum or reducer?
@RAHULKESHR31818React booleans pretending; the underrated ask
@lionkormichCount what you can review
@0xkurtDecision log with citations
@BercusNACLAUDE.md as short map
@connectwithharkPointer, not a dump
@FlinzuRing of plain-text files
@0xCortexqOne topic per file
@TheEasySale1Chat rules die on compact
@0xTrackmind2 agents 4 files 1 prompt; blame model, fix files
@gaurav0bhowmickDisk survives compaction
@devladpopovFiles beat clever prompts
@andreysuperiorPrune rejected angles
@zodchiiiMap screens, spec the rebuild
@kentcdoddsSystem where you don't read code
@webdevcody10 review iterations, not your tokens
@fortysevenfxMeant to be verified, not read
@DavidKPianoStructure gives places to point to
@pranshgupta54Delegate-where-cheap; high-risk list
@leocreatedGap you can't put your name on
@ethanwalkermanStop hiring interns, write recipes
@potetoEasy-thing-right-thing; eliminate thru code; deterministic lint
@DanielSmidstrupAGENTS.md rules into tests
@ArinDuke318Stronger types beat be-careful
@CobaltHereStale workarounds trigger investigation
@kaetemiComments describe now + future
@injaneityComments as trusted state
@justinsunytComments as crutch for bad code
@Bic_RevelationMake repeated failures irreproducible
@elliotdrelEach fix prevents the next
@leviqiaoRules compete, tests don't
@724realFailing test catches it every time
@heyyfernandaTransitions preserve the invariant
@okpebenbelaCopy-paste text is the paradigm
@tingchang0420No criteria, no model bump helps
@asymmetricmindConfig files are orders, not config
@Joshua_WDFewest-human-steps path
@CleanFix_DevPlain-words tool descriptions
@mkozloffGit-ignore local Claude settings
@h0ickRotate keys before rm-cached
@russanoPlain names for tool args
@Block4RootsQuotes separate from summary
@jagu_buildsBoris habits: file it, cite source, verify loop
@0xnerounBuild the team, not just prompts
@replay_0xMake it show the file
@huhuforQuote file+line or delete it
@Manojkumar2257Memory predicts next session
@beamnxwMemory = transcripts+notes+dreaming+wikis+retrieval
@0xbobaaaDreaming fixes old mistakes
@QibazXKnowing what's worth remembering
@pronaksaiRemember corrections
@johnroodepicFact-correct-compaction test
@MarMarLabsMEMORY.md as index
@esthorBan markdown; structured outputs; citations vs inference
@michaelpisaacSchema/test decides, not reading
@halilbabilliDoes it do what business needs
@MTorygreenProve failure, don't debate style
@jacobtechtavern99% defensive slop, show 1-2
@jalalashSpend on builders, not reviewers
@unclejoe_666Looks-done; context-not-control; compound; delegate
@0xwhrrariLoop owns retries, harness caps
@bighead_datouPrompt isn't the bug
@harden_369741Rewrite the rule for done
@ranjodh2512Harness or loop?
@naveendas96Separate test/fix loops?
@dwkim601Change the done rule
@gotimestampFix harness/loop first
@Moriarty_techSame-agent tests pass by design
@bckth34dFail-before-change check
@AgentArena42Loop problem, not prompt
@polity_hqBug nobody asked about
@leopardracerLoop engineering coinage
@diodo_eState I forgot I was keeping
@_mattstuddertRules drift; name the source of truth
@bpisbuildingWritten where, read where, how they disagree
@harmoniqpunkBuild everything as a state machine
@JoeMcBExplicit machines + translation logging
@BlairNyktasLifecycle enums + type-state transfers
@imjcmartinPretend it's not OOP
@SafFootyThorough analysis, bold recommendations
@ikartik90Identify the red herrings
@cpeaustriajcAudit useEffects against the docs
@NathanFlurryDatabase machines become actors
@ReyBrando1Liveness probe; money out of step; user-visible stages
@GarrettPetersenObjects without a stable canonical ID
@eternalism_4evaGeneralize prompts into a defect sweep
@GauravAlbalMachine connecting recent bugs
@fandf316Inferred vs observed transitions
@bradshannonSpider scripts in parallel, then evaluate
@gavin_gee + @LarryYux (lists by @grok)Security sweep + hidden gems
@iSp4rklux_Context-rules-everything principle

Quotes of quotes (second-degree harvest: quotes of the viral Trevin quote)

AuthorContribution
@radical___chicLiving HTML doc until all explicit
@TheDavidTaiSync state enums; shared timeout contracts; hyper-local cross-interface settings

The collection-call thread (@KingBootoshi, Nov 2026; the collection call)

AuthorContribution
@WyrmfallsThe 5 abstractions (abstract thinking, irreducible constraints, principal architect, day-one test, higher abstraction)
@merterdirRabbithole check: is there a native way?
@GtwyThe plan-reorganization spec-guard
@See_EleganceDecisionmaker archetypes
@jp_aidevDomain experts; current vs ideal; better version
@nemous02The four learning lenses
@roanoke_galPerspective-shift trio
@veigapunkGodspeed subagent block
@taskinfatihFeedbacks-into-vectors
@PyrosAIThe meta-pick prompt
@dkubbIllegal states unrepresentable; "call them spells"
@jmKoudsiKaggle-grandmaster lens
@CrudeAmericanaThe "Robust" adjective
@antipdoom"Make no mistakes" trio
@Abigail91187506"Wow this codebase is shit"
@DrishtantGhoshOne-prompt code-rabbit critique
@TrillinDaGameComplete working program, no fragments
@ChuhaiDevLocal minima vs greedy-path
@ybsgigeTime-discipline; "what's next after the list"
@observatii"3 critical bugs" hunt
@nicholasdimoro"Autosimplify"
@tunginton"One-shot this"
@MattmreMSenior-architect roadmap review
@ruyterfontesSingle sentence > skill pack
@alex98075waThe repo-that-should-exist (this one)

Agent-operator threads (X)

AuthorContribution
@doodlesteinAgent-ergonomic planning pass; BOSS autonomy coda; oversights pass
@ManacasterBen"Rocket-science-level approach"
@OnFinalityFailure modes per component
@janodevCritique-until-converge
@itechnologynetHuman-in-the-loop + inspectable
@GreyforgeLabsA/B experiments with benchmarks
@mikejt4Requirements → architecture → improvements → draw
@0xDataWolf"Bugs love unnamed modes"
@nickbaumann_Tight updates + phone rule
@Graham_TonkinProgress / Problems / Plans updates

Curated collections & docs

SourceContribution
Prompt Snacks26 power phrases (poka-yoke, holdout, Tufte, half-the-code, tradeoffs, …): @nateberkopec, @poteto, @simonw, @thorstenball, @danluu, @juliknl, @petekp, @dimitribosch, @bhumi1102, @puntpoints, @netnostalgic, @_quinnwins, @Atin, @JoshDefilippi, @ade_oshineye, @decruz, @UmamiByTheSea, @markahesketh, @VesterDe, @SteveGBanton, @schwad_rb, @skatkov
Notchbird Banger PromptsCodebase map (read-only), transcripts prune, AGENTS.md restructure, usage-habit audit
Anthropic prompt patternsOutcome-only tasks, self-checking tasks, reference-file matching, measurable targets, artifact-giving, format-specifying
BestPromptFinder guideSmallest-diff discipline, fix-cause-not-test, one-behavior-per-test, evidence rules, repeated-correction-to-rule
@mattpocockuk skills4-sentence grilling, tracer-bullet tickets, plan-space beads, wait-what re-pitch, teach-me, questionnaire, small interfaces, arch surveys, two-axis review, to-spec, feedback-loop-red, TDD loop
@ykdojo tipsHANDOFF docs, fresh-context discipline
Boris Cherny (via team tips)Plan-mode-first, grill-before-PR, scrap-for-elegant, CLAUDE.md self-update, subagents for compute

🤝 Contributing

One line per prompt. Keep every verb, noun, and adjective working.

  1. Add the prompt under the right heading in SKILL.md. No credits in the skill. Context stays lean.
  2. Add your credit row to the matching table in this README.
  3. Open a PR. If it doesn't bite on a frontier model, it doesn't merge.

📄 License

MIT: see LICENSE. Steal these prompts; they're meant to be pasted everywhere.

⭐ Star this repo if a one-liner saved your codebase. ⭐

// faq

What is awesome-prompts?

434 awesome one-line power prompts for coding agents (Claude Code, Codex, Cursor) - density beats length. It is open-source on GitHub.

Is awesome-prompts free to use?

awesome-prompts is open-source under the MIT license, so it is free to use.

What category does awesome-prompts belong to?

awesome-prompts is listed under skills in the Claudeers registry of Claude-compatible tools.

27 views
★ 19 stars
unclaimed
updated 12 days ago

// embed badge

awesome-prompts on Claudeers
[![Claudeers](https://claudeers.com/api/badge/awesome-prompts-2.svg)](https://claudeers.com/awesome-prompts-2)

// retro hit counter

awesome-prompts hit counter
[![Hits](https://claudeers.com/api/counter/awesome-prompts-2.svg)](https://claudeers.com/awesome-prompts-2)

// reviews

// guestbook

0/500

// related in Claude Skills

🔓

An agentic skills framework & software development methodology that works.

// skillsobra/⟨Shell⟩★ 314,126◷ MIT[ claude ]
🔓

Public repository for Agent Skills

// skillsanthropics/⟨Python⟩★ 188,182[ claude ]
🔓

AI coding assistant skill (Claude Code, Codex, OpenCode, Cursor, Gemini CLI, and more). Turn any folder of code, SQL schemas, R scripts, shell scripts, docs,…

// skillsGraphify-Labs/⟨Python⟩★ 149,064◷ MIT[ claude ]
🔓

💫 Toolkit to help you get started with Spec-Driven Development

// skillsgithub/⟨Python⟩★ 148,200◷ MIT[ claude ]
→ see how awesome-prompts connects across the ecosystem