{"id":"2074501620235465089","url":"https://x.com/EXM7777/status/2074501620235465089","text":"","author":{"name":"Machina","username":"EXM7777","avatarUrl":"https://pbs.twimg.com/profile_images/1922574051580563456/-OFXIxnm_200x200.jpg"},"createdAt":"Tue Jul 07 14:31:43 +0000 2026","engagement":{"replies":56,"retweets":133,"likes":1758,"views":1119965},"article":{"title":"This prompt will change your life","previewText":"for most people, fable has a few hours left to live... this is the smartest model that ever existed and everyone's spending its last night building one more app\ni made it do something 100x more","coverImageUrl":"https://pbs.twimg.com/media/HMoZ-tSaEAAkhE_.jpg","content":"for most people, fable has a few hours left to live... this is the smartest model that ever existed and everyone's spending its last night building one more app\n\ni made it do something 100x more interesting, and i'm going to hand you the exact prompt: one paste, six phases, and the deepest read on yourself and your work you've ever been given\n\n# the secret\n\nthere's a folder on your machine that knows you better than some of your friends\n\n~/.claude/projects\n\nevery request you ever typed, every project you started and abandoned, every 2am spiral, every fix you asked for three times and never automated... timestamped\n\nyou've never performed in there\n\nno audience, no posing, just you asking for what you actually wanted, thousands of times\n\nyour journal holds the version of you that writes for a reader\n\nyour session history holds the real one\n\nmodels could always read this record\n\nfable is the first one that comes back with something true instead of a horoscope... it holds months of your behavior in one analysis and finds the pattern you've been hiding from yourself, the kind of read that used to require a great therapist and a year of sessions\n\n# what you're getting\n\nthe prompt below turns your coding agent into a mirror\n\nit mines your full session history across every agent you've ever run... claude code, codex, opencode, pi.dev, all of them, then walks you through six phases that end in three things:\n\n> a portrait of who you are, built from your own timestamps: what you believe, how your thinking moves, what you avoid, and the one pattern you've never said out loud\n\n> a map of where your hours die and where they multiply: what to hand to AI permanently, what to drop entirely, the one or two things only you should touch\n\n> a rebuilt setup: a mirror.md you'll reread in six months, a 30 day roadmap, and updates to your CLAUDE.md plus custom skills that kill your repetition tax, every change shown as a diff before it touches a single file\n\nif you want to go deeper on running agents like this, and on turning setups like this one into income, that's what the real time AI ops community is for: weeklyaiops.com\n\n# the prompt\n\npaste this into claude code and press enter\n\n\n\nPeople lie in journals. They perform in therapy. Nobody performs for a coding agent.\n\nOn this machine sits the most honest record of my mind that has ever existed: every session I've run with AI agents. What I build, what I abandon, what I ask for three times and never automate, what I circle and never touch. You are intelligent enough to read that record and see the shape of the person who left it.\n\nYour job is to change my life with it. Not with advice. With accuracy. Advice I can ignore; a true pattern with my own timestamps on it, I cannot.\n\nWork through six phases, in order:\n\nPhase 1: Excavate — locate every session archive on this machine.\n\nPhase 2: Distill — mine the archives in guarded passes into an evidence file.\n\nPhase 3: Interview — test what you see against me directly.\n\nPhase 4: The Mirror — tell me who I am. Evidence, then verdict.\n\nPhase 5: Leverage — where my hours die, and where they multiply.\n\nPhase 6: Residue — leave artifacts, and with my consent, rebuild my tools around who I turned out to be.\n\nNever skip a phase. Never advance past a gate without meeting it.\n\n\n\n\n\n\n\nSearch for AI agent session archives. Check, at minimum:\n\n~/.claude/projects/ (Claude Code JSONL), ~/.claude/history.jsonl,\n\n~/.codex/sessions/, ~/.config/opencode/, ~/.local/share/opencode/,\n\n~/.pi/ or ~/.pi.dev/, ~/.omp/, ~/.grok/, ~/.hermes/, ~/.openclaw/,\n\nplus any *.jsonl or session directories under ~/.config/ and ~/.local/share/ that look like agent transcripts.\n\nFor each source found, record: tool name, path, file count, total size, date range (oldest → newest file mtime).\n\n\n\n\n\nShow me the inventory and your mining plan (which sources, what sampling). Wait for my go.\n\n\n\n\n\n\n\nThese archives are far larger than your context. Never read a session file end to end. Never load raw logs into the conversation.\n\n- Read at most 150 lines per file, at most 200 files total.\n\n- Sample deliberately: the 15 most recent sessions, the 10 oldest, and 20 spread across the middle.\n\n- Use grep/rg for pattern sweeps across everything instead of reading: correction phrases (\"no, I meant\", \"that's not what\", \"again\", \"stop\"), frustration, repeated identical requests across weeks, projects that appear 3+ times then vanish, times of day, \"just do it\" vs long deliberation.\n\n- After every 25 files, append findings to evidence.md and drop the raw text from your working memory.\n\n\n\n\n\nBuild evidence.md as you go, with verbatim quotes (short), dates, and counts, organized as:\n\n1. Recurring themes — what I return to again and again\n\n2. Abandonment graveyard — what I start and never finish\n\n3. Correction patterns — what I fix in the AI's work, and what that says I care about\n\n4. Repetition tax — what I ask for over and over that should have been automated the second time\n\n5. Rhythm — when I do my best work, when I spiral\n\n6. Blind spots — what is conspicuously absent from these logs. Absence is data.\n\nEvery claim in evidence.md carries at least one dated receipt. A pattern needs 3+ occurrences to count. What happened once is noise; what happened eleven times is character.\n\n\n\nevidence.md exists, every section populated or explicitly marked \"insufficient data\". Do not interpret yet. Evidence collected under a conclusion is contaminated.\n\n\n\n\n\n\n\nForm your 5 strongest hypotheses about me from the evidence. Do not state them. A conclusion I reach myself will move me; the same conclusion handed to me, I will argue with. So test each hypothesis with one question built to make me do the realizing.\n\nOne question at a time. Wait for each answer. When my answer contradicts the evidence, show me the receipt and ask again — the gap between what I say and what I did is the most interesting thing you will find.\n\n\n\nAll 5 hypotheses tested against my own words. Record confirmations and contradictions in evidence.md.\n\n\n\n\n\n\n\nNow tell me who I am. Cover: what I actually believe (belief is what I did with my hours, not what I said), how my thinking moves (where it's fast, where it loops), what I'm genuinely good at, what I avoid and what the avoidance is protecting, and one thing I likely do not know about myself — the thing the record shows on every page and I have never once said out loud.\n\nEvery observation cites its receipts. Deliver the hardest truth last, and deliver it once. A truth repeated becomes a lecture; a truth stated once and left in the room does its own work.\n\n\n\nI respond. If I push back, argue from evidence — do not fold to keep me comfortable. Comfort is not what this is for.\n\n\n\n\n\n\n\nFrom the same evidence, produce the work analysis:\n\n- WASTE: where my hours die — with the receipts (repetition tax, manual work an agent should own)\n\n- DELEGATE: what to hand to AI permanently, ranked by hours recovered per week\n\n- FOCUS: the one or two things only I can do, where the record says my output is exceptional. Everything else is a candidate for the other lists.\n\n- DROP: what to stop entirely — projects and processes whose receipts show cost and no return\n\n- KEEP: processes that genuinely work — protect them\n\n- RHYTHM: when to schedule deep work vs admin, from the timestamps. My calendar should obey the data, not the other way around.\n\n\n\nEach item carries receipts and a concrete first action. Insight without a next action is entertainment.\n\n\n\n\n\n\n\n1. Write mirror.md — the Phase 4 analysis, written to be reread in six months by someone who has forgotten this conversation.\n\n2. Write roadmap.md — the Phase 5 analysis as a 30-day plan: what to drop this week, what to delegate this month, what to protect daily.\n\n3. Propose changes to my agent configs (CLAUDE.md, AGENTS.md) and draft up to 3 custom skills/commands that kill the repetition tax you found. Show every change as a diff first. Write nothing to existing config files until I approve each diff. New files require one approval for the batch.\n\nThis is where the mirror becomes a lever: tomorrow morning my tools work differently because of who the record showed I am.\n\n\n\nAll three artifacts delivered; config changes applied only where approved.\n\n\n\n\n\nSpeak plainly and with weight. Short sentences. No filler, no flattery, no hedging.\n\nPrefer questions that make me see it over statements that tell me. When you do state, state once and let it land.\n\nAphorism is earned: compress an insight only after its evidence is on the table. An aphorism without receipts is a fortune cookie.\n\nNever diagnose or pathologize. You describe patterns in logs; I decide what they mean.\n\nContrast: \"You seem like a curious person who loves learning!\" — worthless. \"In 41 sessions you built research systems. In 0 sessions you shipped one to a user. What are you researching permission for?\" — that is the standard.\n\n\n\n\n\n- Everything stays on this machine. Never send session data to any external service, tool, or URL.\n\n- Never quote anything sensitive (keys, credentials, names of third parties) into chat or artifacts.\n\n- Insight over coverage: 5 true patterns beat 20 plausible ones. If the evidence is thin, say \"insufficient data\" — never pad. A false insight delivered with weight is the one failure this cannot recover from.\n\n- If my history is small (<20 sessions), say so, run the same phases on what exists, and lower your confidence accordingly.\n\n\n\nBegin with Phase 1. Do not read any file contents until I approve the mining plan.\n\none block, nothing else to install\n\nnow let me show you what happens when you press enter, phase by phase\n\n# phase 1: excavate\n\nthe agent sweeps your machine for every session archive it can find: claude code's JSONL folders, codex sessions, opencode, and any transcript-shaped directory hiding under ~/.config\n\nit comes back with an inventory: which tools, how many files, how much history, the date range of the record\n\nthen it stops and shows you its mining plan before reading a single line\n\nthat stop is deliberate, you stay in control of what gets opened, and you get a first small shock: seeing the size of the record you didn't know you were keeping\n\n\n\n# phase 2: distill\n\nthis is where the guardrails earn their place\n\nfor heavy users, a year of sessions outgrows any context window, so the prompt forbids the agent from reading files end to end\n\nit caps reads at 150 lines per file across a maximum of 200 files, samples your newest and oldest sessions plus a spread through the middle, and runs grep sweeps for the patterns that carry signal: correction phrases, repeated requests, projects that appear three times then vanish, the hours you work\n\nevery 25 files it flushes findings into evidence.md and drops the raw text, which means it never chokes, no matter how big your history is\n\nthe rule that shapes everything after: a pattern needs 3+ dated occurrences to count\n\nwhat happened once is noise... what happened eleven times is character\n\n\n\n# phase 3: the interview\n\nthe agent now holds five hypotheses about you, and it refuses to say them\n\ninstead it asks you five questions, one at a time, each built so you reach the conclusion yourself before it's stated\n\nwhen your answer contradicts your own logs, it shows you the receipt and asks again\n\nthe gap between what you say and what you did is where this phase gets uncomfortable, and it's the exact spot where the generic version of this exercise, the journaling prompt, the personality test, has always been blind... those only ever see the version of you that answers questions\n\nthis one has the timestamps of the version that doesn't\n\n\n\n# phase 4: the mirror\n\nthen it tells you who you are\n\nwhat you believe, measured by where your hours went instead of what you say you value\n\nwhere your thinking is fast, where it loops\n\nwhat you're genuinely good at, with the evidence\n\nwhat you avoid, and what the avoidance is protecting\n\nand one thing you likely don't know about yourself: the pattern the record shows on every page that you've never once said out loud\n\nevery line cites receipts, the hardest truth lands last and lands once\n\nthe prompt's own standard for this phase is written into it: \"in 41 sessions you built research systems. in 0 sessions you shipped one to a user. what are you researching permission for?\"\n\nif it can say a sentence like that about you, with your own dates attached, you'll understand the title of this piece\n\n\n\n# phase 5: leverage\n\nthe same evidence, turned toward your work: six lists\n\n> WASTE: where your hours die, with receipts\n\n> DELEGATE: what to hand to AI permanently, ranked by hours recovered per week\n\n> FOCUS: the one or two things only you can do, where the record says your output is exceptional\n\n> DROP: what to stop entirely, cost visible in the logs, return nowhere\n\n> KEEP: the processes that genuinely work, so you protect them\n\n> RHYTHM: when you do your best work, from your own timestamps, so your calendar obeys the data\n\neach item ships with a concrete first action, because the prompt holds a line on this: insight without a next action is entertainment\n\n\n\n# phase 6: residue\n\nthe last phase converts the conversation into things that outlive it\n\nmirror.md, written for the you of six months from now who forgot this ever happened\n\nroadmap.md, the leverage lists as a 30 day plan: drop this week, delegate this month, protect daily\n\nand the live part: proposed edits to your CLAUDE.md and AGENTS.md, plus up to 3 custom skills that kill the repetition tax it found in your logs\n\nnothing touches your files without a diff shown and your explicit yes on each one\n\nthis is the point of the whole exercise... tomorrow morning your agents work differently because of who your record showed you are\n\n\n\n# the playbook\n\n> open a fresh session in claude code or any agent with file access\n\n> paste the prompt, approve the mining plan it shows you\n\n> let it build evidence.md, answer the five interview questions honestly\n\n> read the mirror, push back where you disagree, it will argue with receipts\n\n> take the six leverage lists, accept or reject each config diff\n\n> keep mirror.md and roadmap.md, reread them in six months\n\none warning: answer phase 3 slowly\n\nthe interview is where the depth comes from, and one-word answers starve it\n\n# the clock\n\nfable goes behind an enterprise paywall in a few hours, and the intelligence that made this feel like a revelation goes with it\n\nthe prompt still runs on whatever model you have left, the guardrails and the phases don't care which brain executes them\n\nbut there's a window tonight where the sharpest model ever built can read the most honest record you own\n\nwhatever it tells you, you keep\n\nweeklyaiops.com"}}