
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.
# copy the skill dir into your project: # .claude/skills/awesome-prompts/ (or ~/.claude/skills/awesome-prompts/ for all projects)
git clone https://github.com/Drix10/awesome-prompts
// compatibility
| Platforms | cli, api, web |
|---|---|
| Operating systems | — |
| AI compatibility | claude |
| License | MIT |
| Pricing | open-source |
| Language | — |
⚡ 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.
| Part | Count | What it steers |
|---|---|---|
| 🧠 Meta | 3 + rule | Let the agent pick its own prompts |
| 🏛 Architecture | 22 | Abstraction-hunting, system framing |
| 🔍 State-machine audits | 63 | The complete implicit-state-machine playbook |
| 🦄 Unicorn hunts | 19 | DISCOVER hunts: invariants, transactions, trust |
| 🐛 Debugging | 66 | Symptoms vs roots, failure modes, hunts |
| 📐 Planning | 48 | Specs, grilling, handoffs, autonomy |
| ✨ Quality & repair | 67 | Smaller diffs, TDD, repair sets, security sweep |
| ✅ Verification | 35 | Evidence, holdouts, PROVE set, ship decisions |
| 👀 Review | 23 | Two-axis reviews, PR shape |
| 🧹 Hygiene | 3 | Transcripts, AGENTS.md, usage audits |
| 📚 Learning | 15 | Mental models, diagrams, ELI5 |
| 🎛 Steering | 70 | Morale, BOSS coda, updates, skills |
| 📝 Notes | principles | Field 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")
| Author | Contribution |
|---|---|
| @trevin | The expanded one-file audit (states, invariants, transition tables, retries + concurrency, prioritize, group, fix, keep updated) |
| @Nikzaaaaad | The 9 follow-ups (false screens, UI business logic, drifted copies, unenforced DB rules, races, swallowed errors, magic numbers, loose types, effects-as-events) |
| @flowforth_76 | The full phased audit (discovery → verification → plan → implement) |
| @carpenterdroid | The refactor-directive mega-prompt (eliminate state, derive values, explicit machines, regression + integration checks) |
| @asmirkn | Transition tables; impossible-but-reachable transitions |
| @aloobhujiyan | Where 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_666 | Concrete event sequences; permitted vs reachable states |
| @eclarkhalid | Cite handlers + conditions; mark proposed changes separately |
| @akashc777 | Hiding spots: flags the UI never designed for |
| @jj_kovacs | 16 states on paper, 5 real; bugs in the other 11 |
| @kabelsalat_info | List combos that can't happen but code allows |
| @Talljackcv | Cancelled must never return to success |
| @BongJupite43260 | Cross-component inconsistent paths |
| @sathya_partha | Screen rules no component enforces |
| @gippp69 | Harness rules: allow/ask/deny, BLOCKED, receipts |
| @xsmryze | Each model only on its best job |
| @jasonmello | Plan in 5 bullets before anything |
| Amin Roosta | Call a spade a spade |
| John Rush | Fresh chat after loops |
| @Voxyz_ai | Rate bugs 1-10; no designing for rare cases |
| @_____Calum | Narrow beats smart; hints make it worse |
| @franzejr | Tickets need behavior, constraints, checks |
| @MisterSnowCrash | Regression test failing-then-passing artifact |
| @BlossomAILabs | Reviewer gets acceptance criteria first |
| @huesofbanter | Test for removing workarounds |
| @cnwenf | Checkable handoff results |
| @itsvdruzko | Two agents disagree → edit ticket |
| @z1njo | Interaction, not a picture |
| @choblin29 | Requests vs hard boundaries |
| @kr4chinin | Visualize the state flow |
| @sjmurdoch | Verify correctness in Quint |
| @garbnzgh | Subagents + propose building it into reviews/workflows |
| @badbirb | Frame out stateful elements and plan before coding |
| @EternitiesAI | Failing regression test = boundary of suspected vs fixed |
| @Pps831 | Global-rules prevention; no-implicit-state-machines rule |
| @feddiwhip | Unenforced invariants |
| @bonsaielectric | Implicit invariants |
| @peytoncasper | Non-terminal states variant |
| @jussisaur | The explicit inversion (turso) |
| @twid | "Find all tautologies" |
| @wahome_thuku | The legal port: implicit obligations in contracts |
| @jimlux | "Find the dumb requirements" (MCAS debate) |
| @cocobiskit | Adversarial red-teaming |
| @joshest3p | "Pushing determinism down" |
| @krimuthu | Autonomous 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" |
| @Shannoncode | Root-cause tokens; SDK/backend/schema context assembly |
| @menhguin | Scientific method; decision frameworks (Chesterton, OODA, autonomy levels) |
| @matt_dot_sol | Targeted "are you sure" double-checks |
| @PreciousBa82157 | "If it's not in the prompt, it doesn't exist" |
| @bigdatachads | Research-first, then root-cause |
| @asavs_ | Roleplay user / agent / system internals |
| @adityavg13 | Internalize metaphors (allegory of the cave) |
| @impervium888 | Explain it as a story |
| @codeby_ABI | Recursive prompt booster |
| @0xRajat | One-liners over elaborate skills |
| @ZazenCodes | Doing things wrong vs doing the wrong things |
| @teodorio | Master-agent routing; vector-path steering |
| @PawelPSzczesny | Mine operators, combine into new prompts |
| @BadGatewayCo | The combo: machines + invariants + deepen modules |
| @webtkdev | Convert to actors; rules-file ban |
| @mrcommando712 | Map states before fixing; late-response check |
| @PaavoKarlin | Report shape: found / decide / bugs / sound |
| @imahmood | Interview-me + Simplified Technical English |
| @cryptoquick / @GarrettPetersen | Explicit, tested, documented |
| @feddiwhip | The 16 unicorn hunts + PROVE/REPAIR sets + evidence prompt |
| @amronos / sprocket | Cleanup sweep: over-engineering, dead text, duplication, coverageless tests |
| @GKev1n | Read-only guardrail: states, transitions, conflicts + line numbers; wild boolean machines |
| @buchmanster | Specs first, then tests from specs |
| @oderajoseph | Scope the first refactor |
| @Nikzaaaaad | 9-more list: false screens, UI logic, DB rules, magic numbers, loose types |
| @VincentLinwx | Request IDs + cancellation vs state |
| @maxxrubin_ | Blast-radius thinking per change |
| @psychic_jammer | Ban comments except the exceptional |
| @Lion_GameDev | Unmentioned edge cases get assumed |
| @charleshxpe | Structured scan: states, moves, conflicts, risk rank |
| @0xUltraInstinct | Make it permanent in AGENTS.md |
| @literallyhimmmm | Read the bugs first; half aren't real |
| @alexlee20122022 | Walk the unhappy paths principle |
| @kr4chinin | Leads-not-findings; Confirmed/Likely/Smell; recreate-not-reset |
| @KelvinW_info | One named principle per run |
| @deepman_1 | Long-running things become explicit machines |
| @ddyods | Rules in the harness, practices in the prompt |
| @jbaruch | Right 300 tokens beat 100k noisy ones |
| @yanggao97566 | Machines disagreeing about done |
| @boringdev77 | Flags or status strings? |
| @CarbonMatchAI | Vocabulary is a senior skill |
| @IamYashKapoor | Booleans secretly meaning step 3 |
| @arpittalksstuff | isReallyLoading and isDoneForReal |
| @asmirkn | Enum + transition table + failing test |
| @fishjar | Unnamed state bits |
| @PowerduckLabs | Retries plus flags; name and reject |
| @jeffbruchado | Handler order is the only spec |
| @OdysseyStack | The unnamed isLoading conjunction |
| @johnroodepic | A type that rejects the fourth flag |
| @looking5452 | Model with typestate |
| @hboon | Study-and-find phrasing; 16 short prompts (elaborate simply, root cause, supply-list, essence, mock, review-context) |
| @illulachy_ | Enum or reducer? |
| @RAHULKESHR31818 | React booleans pretending; the underrated ask |
| @lionkormich | Count what you can review |
| @0xkurt | Decision log with citations |
| @BercusNA | CLAUDE.md as short map |
| @connectwithhark | Pointer, not a dump |
| @Flinzu | Ring of plain-text files |
| @0xCortexq | One topic per file |
| @TheEasySale1 | Chat rules die on compact |
| @0xTrackmind | 2 agents 4 files 1 prompt; blame model, fix files |
| @gaurav0bhowmick | Disk survives compaction |
| @devladpopov | Files beat clever prompts |
| @andreysuperior | Prune rejected angles |
| @zodchiii | Map screens, spec the rebuild |
| @kentcdodds | System where you don't read code |
| @webdevcody | 10 review iterations, not your tokens |
| @fortysevenfx | Meant to be verified, not read |
| @DavidKPiano | Structure gives places to point to |
| @pranshgupta54 | Delegate-where-cheap; high-risk list |
| @leocreated | Gap you can't put your name on |
| @ethanwalkerman | Stop hiring interns, write recipes |
| @poteto | Easy-thing-right-thing; eliminate thru code; deterministic lint |
| @DanielSmidstrup | AGENTS.md rules into tests |
| @ArinDuke318 | Stronger types beat be-careful |
| @CobaltHere | Stale workarounds trigger investigation |
| @kaetemi | Comments describe now + future |
| @injaneity | Comments as trusted state |
| @justinsunyt | Comments as crutch for bad code |
| @Bic_Revelation | Make repeated failures irreproducible |
| @elliotdrel | Each fix prevents the next |
| @leviqiao | Rules compete, tests don't |
| @724real | Failing test catches it every time |
| @heyyfernanda | Transitions preserve the invariant |
| @okpebenbela | Copy-paste text is the paradigm |
| @tingchang0420 | No criteria, no model bump helps |
| @asymmetricmind | Config files are orders, not config |
| @Joshua_WD | Fewest-human-steps path |
| @CleanFix_Dev | Plain-words tool descriptions |
| @mkozloff | Git-ignore local Claude settings |
| @h0ick | Rotate keys before rm-cached |
| @russano | Plain names for tool args |
| @Block4Roots | Quotes separate from summary |
| @jagu_builds | Boris habits: file it, cite source, verify loop |
| @0xneroun | Build the team, not just prompts |
| @replay_0x | Make it show the file |
| @huhufor | Quote file+line or delete it |
| @Manojkumar2257 | Memory predicts next session |
| @beamnxw | Memory = transcripts+notes+dreaming+wikis+retrieval |
| @0xbobaaa | Dreaming fixes old mistakes |
| @QibazX | Knowing what's worth remembering |
| @pronaksai | Remember corrections |
| @johnroodepic | Fact-correct-compaction test |
| @MarMarLabs | MEMORY.md as index |
| @esthor | Ban markdown; structured outputs; citations vs inference |
| @michaelpisaac | Schema/test decides, not reading |
| @halilbabilli | Does it do what business needs |
| @MTorygreen | Prove failure, don't debate style |
| @jacobtechtavern | 99% defensive slop, show 1-2 |
| @jalalash | Spend on builders, not reviewers |
| @unclejoe_666 | Looks-done; context-not-control; compound; delegate |
| @0xwhrrari | Loop owns retries, harness caps |
| @bighead_datou | Prompt isn't the bug |
| @harden_369741 | Rewrite the rule for done |
| @ranjodh2512 | Harness or loop? |
| @naveendas96 | Separate test/fix loops? |
| @dwkim601 | Change the done rule |
| @gotimestamp | Fix harness/loop first |
| @Moriarty_tech | Same-agent tests pass by design |
| @bckth34d | Fail-before-change check |
| @AgentArena42 | Loop problem, not prompt |
| @polity_hq | Bug nobody asked about |
| @leopardracer | Loop engineering coinage |
| @diodo_e | State I forgot I was keeping |
| @_mattstuddert | Rules drift; name the source of truth |
| @bpisbuilding | Written where, read where, how they disagree |
| @harmoniqpunk | Build everything as a state machine |
| @JoeMcB | Explicit machines + translation logging |
| @BlairNyktas | Lifecycle enums + type-state transfers |
| @imjcmartin | Pretend it's not OOP |
| @SafFooty | Thorough analysis, bold recommendations |
| @ikartik90 | Identify the red herrings |
| @cpeaustriajc | Audit useEffects against the docs |
| @NathanFlurry | Database machines become actors |
| @ReyBrando1 | Liveness probe; money out of step; user-visible stages |
| @GarrettPetersen | Objects without a stable canonical ID |
| @eternalism_4eva | Generalize prompts into a defect sweep |
| @GauravAlbal | Machine connecting recent bugs |
| @fandf316 | Inferred vs observed transitions |
| @bradshannon | Spider 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)
| Author | Contribution |
|---|---|
| @radical___chic | Living HTML doc until all explicit |
| @TheDavidTai | Sync state enums; shared timeout contracts; hyper-local cross-interface settings |
The collection-call thread (@KingBootoshi, Nov 2026; the collection call)
| Author | Contribution |
|---|---|
| @Wyrmfalls | The 5 abstractions (abstract thinking, irreducible constraints, principal architect, day-one test, higher abstraction) |
| @merterdir | Rabbithole check: is there a native way? |
| @Gtwy | The plan-reorganization spec-guard |
| @See_Elegance | Decisionmaker archetypes |
| @jp_aidev | Domain experts; current vs ideal; better version |
| @nemous02 | The four learning lenses |
| @roanoke_gal | Perspective-shift trio |
| @veigapunk | Godspeed subagent block |
| @taskinfatih | Feedbacks-into-vectors |
| @PyrosAI | The meta-pick prompt |
| @dkubb | Illegal states unrepresentable; "call them spells" |
| @jmKoudsi | Kaggle-grandmaster lens |
| @CrudeAmericana | The "Robust" adjective |
| @antipdoom | "Make no mistakes" trio |
| @Abigail91187506 | "Wow this codebase is shit" |
| @DrishtantGhosh | One-prompt code-rabbit critique |
| @TrillinDaGame | Complete working program, no fragments |
| @ChuhaiDev | Local minima vs greedy-path |
| @ybsgige | Time-discipline; "what's next after the list" |
| @observatii | "3 critical bugs" hunt |
| @nicholasdimoro | "Autosimplify" |
| @tunginton | "One-shot this" |
| @MattmreM | Senior-architect roadmap review |
| @ruyterfontes | Single sentence > skill pack |
| @alex98075wa | The repo-that-should-exist (this one) |
Agent-operator threads (X)
| Author | Contribution |
|---|---|
| @doodlestein | Agent-ergonomic planning pass; BOSS autonomy coda; oversights pass |
| @ManacasterBen | "Rocket-science-level approach" |
| @OnFinality | Failure modes per component |
| @janodev | Critique-until-converge |
| @itechnologynet | Human-in-the-loop + inspectable |
| @GreyforgeLabs | A/B experiments with benchmarks |
| @mikejt4 | Requirements → architecture → improvements → draw |
| @0xDataWolf | "Bugs love unnamed modes" |
| @nickbaumann_ | Tight updates + phone rule |
| @Graham_Tonkin | Progress / Problems / Plans updates |
Curated collections & docs
| Source | Contribution |
|---|---|
| Prompt Snacks | 26 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 Prompts | Codebase map (read-only), transcripts prune, AGENTS.md restructure, usage-habit audit |
| Anthropic prompt patterns | Outcome-only tasks, self-checking tasks, reference-file matching, measurable targets, artifact-giving, format-specifying |
| BestPromptFinder guide | Smallest-diff discipline, fix-cause-not-test, one-behavior-per-test, evidence rules, repeated-correction-to-rule |
| @mattpocockuk skills | 4-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 tips | HANDOFF 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.
- Add the prompt under the right heading in
SKILL.md. No credits in the skill. Context stays lean. - Add your credit row to the matching table in this README.
- 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.
// embed badge
[](https://claudeers.com/awesome-prompts-2)
// retro hit counter
[](https://claudeers.com/awesome-prompts-2)
// reviews
// guestbook
// related in Claude Skills
An agentic skills framework & software development methodology that works.
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,…
💫 Toolkit to help you get started with Spec-Driven Development