Problem Mining Loop: real complaints, graded and counted, before you build anything
by PaperChaseWebb
Find out what one kind of buyer complains about, in their own words. Every complaint is a row with the link, the date posted, the exact quote and a salted key in place of the poster's handle, graded gripe, pain, budget or buyer. Clusters count independent posters, not posts, and leave out prompted answers and your own accounts. A cluster is proposed only when enough people at pain or higher, from more than one source, say so and one shows money. Public pages only, robots.txt and terms checked, GET only, and nobody in the community is ever contacted. Proposals wait for your yes at your own terminal; fifteen read-only checks run on the set.
- Price
- $19
- Difficulty
- Intermediate
- Setup
- 1 hr
- Format
- .md file
problem-mining-loop-real-complaints-graded-and-counted.md · 81 KB · instant download after checkout
Want this built for you instead? We can build it for you.
What you end up with
A checked log of what your buyers said, a one-page report of which problems are shared and paid for, and next steps that happen only after your yes.
What you get
- 19 sections in one .md file
- The loop folder: loop.json, sources, complaints, clusters and proposals, with blank templates
- Rules for sources (public, robots.txt, terms with a date), verbatim quotes, the four grades, solicited answers, clusters and the propose rule
- mine.py: set hash, salted poster keys, gate (exit codes 0, 11, 12), a read-only GET fetcher that obeys robots.txt and stops at logins, redirects and rate limits, tally, a one-page report, owner approval tied to a sha256, verify, and one use per yes
- mine_check.py: fifteen read-only checks, with a seeded test set (15 findings, then 0)
- A skill file for Claude Code, Codex and Hermes, an AGENTS.md snippet and a Grok Bot rules block
- Ten prompts, a failure table and a worked example
- A look at pain-point tools, prompt kits and research passes, with sources
Who it is for
Anyone who wants the next product or offer to rest on what buyers actually said, with links, rather than on a list of problems that sound right.
Works with
Solid chips: the agent is named in this listing's own tools list. Dashed chip: A skill file for Claude Code, Codex and Hermes, an AGENTS.md snippet and a Grok Bot rules block; the scripts were tested on Linux with Python 3.13, the agent files were not loaded inside each harness.
Tools needed
Cost to run: The pack is the only purchase. The scripts use no model tokens; the agent that reads and grades uses your own plan.
Time to result: A first graded log and report in an afternoon of reading
How it works
Run the tests
Copy the problem-mining-loop folder and run sh test_all.sh. Nothing in it touches the internet; the fetch tests use a local server.
Expected: 15 findings on the seeded set and 0 on the fixed one, 15 of 15 single faults caught alone, checker tests 43 passed, mine.py tests 53 passed, exit 2 on a missing folder.
Set up the loop and its sources
Run mine.py init, fill loop.json with the owner, one buyer type and the question, then list sources with prompts 1 and 2, checking robots.txt and terms for each.
Expected: sources.csv with public sources marked allowed, login-only ones left out with allowed no, and a terms check date on each.
Record, grade and cluster
Use prompts 3 to 7 to record complaints word for word with poster keys, grade them, cluster by workflow, run mine.py tally and decide watch, drop or propose.
Expected: complaints.csv and clusters.csv where the counts match the tally and every propose cluster clears the rule.
Check, report and approve
Run mine_check.py until only P15 is left, then mine.py report. The owner runs mine.py approve DIR --set, then --proposals for the ones they agree with, at their own terminal.
Expected: report.md with clusters, quotes and proposals, and mine.py verify passing once per approved proposal.
Sample prompts, free to use
For each post you listed, add a row to complaints.csv: the exact words of the complaint up to 280 characters, verbatim yes, the poster key from mine.py key (never the handle), the source, the link and dates, and solicited yes if it answers a question asking for complaints. Do not paraphrase. Grade each new complaint gripe, pain, budget or buyer. For pain, write in tried what they tried. For budget and buyer, copy the evidence words from the quote. When unsure, grade lower and say which ones you were unsure about.
Prompts 4 and 5 of 10.
For each cluster, apply the rule in loop.json and propose watch, drop or propose, with a one-line next step. For each propose cluster, write one or two proposals for my yes. Do not contact anyone.
Prompt 7 of 10.
Free preview of the file
# Problem Mining Loop: real complaints, graded and counted, before you build anything Version 1.0, 2026-10-29. By PaperChaseWebb. A method and a working loop for finding out what one kind of buyer complains about, in their own words, and whether any of them already pay to make it stop. Five plain files, a script that fetches public pages read-only, keys posters without storing who they are, tallies clusters and records the owner's yes, a read-only checker with fifteen checks, a one-page report, templates, a seeded test set, ten prompts and the tests, all Python 3 standard library. Works with Claude Code, Codex and Hermes through a skill file and an AGENTS.md snippet, and with Grok Bot through a rules block. Tested as described in section 11: everything ran on Linux with Python 3.13, including owner approvals typed at a pseudo-terminal and the fetch rules against a local test server. The skill file, snippet and rules block were not loaded inside the harnesses for this release, and the loop has not been run against a live forum or review site. Each is marked where it matters. Most "pain point research" ends as a list of problems that sound right. Some of them were written by a model and never said by anyone. Some were said once, by one loud poster, eleven times. Some were answers to a thread the researcher started, which means they were prompted. And a few were scraped from places whose terms say not to. The list then turns into a product, and the product meets buyers who nev
Test record
- Date
- Oct 29, 2026
- What was run
- mine_check.py on a seeded set and a fixed copy, each planted fault alone, a missing input, and the unit tests. Linux, Python 3.13.
- Result
- 15 findings on the seeded set, 0 on the fixed one; 15 of 15 planted faults caught alone; exit 2 on missing input; nothing written. 43 checker and 53 mine.py tests passed, 0 failed; fetch was tested against a local server.
- Not run
- The agent files inside Claude Code, Codex, Hermes or Grok Bot. Fetching a live site. Windows.
Copied from this item's dated test notes. What was not run stays in.
Worth knowing
- mine.py records, counts and gates; it does not read pages for meaning or pick quotes, and it cannot stop an agent that never asks. This is not a sandbox.
- The skill file, AGENTS.md snippet and Grok Bot rules block were not loaded inside Claude Code, Codex, Hermes or Grok Bot for this release, so how well each harness grades or keeps quotes verbatim was not measured. mine.py fetch was tested only against a local server, not a live site. Whether a site's terms allow your use is your judgement, recorded with a date. Poster keys keep handles out of the files; they are not anonymity. The personal details check catches common shapes, not every one. The test business is made up. Linux only; Windows was not tested.
- A cluster that clears the propose rule is a reason to test an offer, not proof that it will sell.
- The approval step refuses piped input and the environment variables Claude Code, Codex, Hermes and Grok Bot set. That stops an agent approving by mistake; it is not a lock against someone with the same account.
- On 2026-10-29, on Linux with Python 3.13: mine_check.py gave 15 findings on the seeded set (each check once) and 0 on the fixed one, caught 15 of 15 single faults with only its own check, wrote nothing, and exited 2 on a missing folder, a missing complaints.csv or broken loop.json; checker tests 43 passed, 0 failed; mine.py tests (fetch against a local server, approvals at a pseudo-terminal) 53 passed, 0 failed.
Questions
Is this a Reddit scraper?
No. It works with any public place your buyers write, and it has no scraper. mine.py fetch sends one GET for one page on a source you allowed, after checking robots.txt, and stops at logins, redirects and rate limits. Where a site's terms do not allow fetching, read by hand or use its official API on its terms.
Will my agent post or reply in the community?
No. post, reply, dm, vote, seed and login are gated never, the checker fails a loop that gates them otherwise, and complaints from your own accounts or threads you started are flagged.
Why count posters instead of posts?
One person posting eleven times is one person. Posters are counted by a salted key, so repeat posts count once and the handle never goes in the file.
I own Workflow, Not Wrapper. Do I need this?
Its section on problem mining is a one-page method with three grades. This pack is that step as a monthly loop with four grades, poster keys, a propose rule, a read-only fetcher, fifteen checks and the owner's yes.
Details
Version 1.0 · Updated Oct 29, 2026 · Markdown version of this page
License: Use on your own business and to run research for your clients. Do not resell or republish the pack itself. Download terms
Related
Problem Mining Loop: five rules and a five-check script (free sample)
From the Problem Mining Loop: five rules for a complaint log you can trust (one owner and one buyer type, a link and date on every complaint, quote never paraphrase, money shown not guessed, count people not posts), and a read-only script that checks one log for them.
Research Contract for Agents: claims, sources and dates your agent must keep
A contract your agent follows on every research run, kept as plain files: each claim with its source, date and confidence, primary sources before secondary, quotes word for word with a location, contradictions listed, unknown allowed, and no citation to anything it did not open. A fifteen-check script reads the folder and names every break in a plain sentence, and nothing is published or acted on until the owner approves the exact report by its sha256.
Check the Claim (expanded edition): a method, prompts and a skill for checking AI claims
Make your agent check income, tool, popularity, quote, news and offer claims before they reach a brief, a pitch, a purchase or a post: label how much was read, run the creator's own arithmetic, trace sources back to first hand, and give each claim one of five verdicts on a card the owner approves. v1.1 adds an evidence level label (full, caption, summary or wall), four new arithmetic tests, a section on tracing a source back to first hand, screens for offers, quotes and slogans, four more red flags, a sixth worked example, a claim folder with a fourteen-check read-only checker and an owner approval helper, agent files for Grok Bot, Claude Code, Codex and Hermes, and twelve new prompts (26 in all). Price unchanged.
Workflow, Not Wrapper: test an AI product idea, shape the offer and set the price
A prompt pack and method for anyone building or selling on top of a model. A seven-score absorption test, a one-page workflow product spec, a demo gap check, a problem mining loop, a five-talk buyer script, a channel translation table, a model column check, a true cost rebuild at three usage levels, an offer ladder with pricing rules, a price anchor audit, a check for copied success stories, a fidelity gate for generated output, a failure table, a worked example, an idea folder with a tested checker, a rules block and 19 prompts. What changed in v1.1: the demo gap check, buyer talks, true cost, price anchors and the checker are new.
Compile Your Research: a cited, checked second brain your agents keep
Your agent compiles saved sources into linked markdown pages where every fact cites its file, conflicts get their own page, and nothing is the record until a person ratifies it. Four folders, ten rules, seven page types, ten prompts, a health check script that needs no AI, and a seeded test vault with nine planted problems.
Agency Install Kit: install one agent workflow in two weeks, with a binder a script can check
Five paid packs and the free Readiness Self-Check, tied into one two-week install by a new Install Binder: plain files for readiness, one workflow, agent owners and tested pauses, a gate register, a reach inventory, a cost cap, a dry run with refusal cases, a scorecard with kill rules and a run sheet with an off switch. A fifteen-check script names every missing decision, and the owner's go-live yes is tied to the binder's sha256.
Agent Readiness Self-Check: score one workflow before you hand it to an AI agent
A free self-check for agencies and small businesses. Name the decision, then score one workflow across eight areas (workflow mapped, data access, approval gates, owner per agent, handoffs, cost tracking, security and vetting, measurement) with an evidence rule, three gate rules and a skeptical pass, and read your band and the next steps it allows. v1.1 starts with the decision the score will change, scores unknowns as unknown, adds rules for conflicting facts and lookalike public records, ties every band to the next steps it allows with the Production Readiness Audit before go-live, makes owner approval concrete (prepared and approved status, a one-reply approval line, decisions the agent must leave open, a done test), adds agent files for Grok Bot, Claude Code, Codex and Hermes, a second worked example (missed call to booked job), a readiness folder with a fifteen-check read-only checker, a front door test with four questions to ask anyone selling you an agent install, and a plain note on how it was tested. Still free.
Source Shelf: a short, owner-approved list of industry reports your team reads, and a way to read them
Keep a small set of major industry reports as your team's default evidence base: each with its edition, publish date, link on the publisher's own site, the date someone last looked for a newer edition, and a note on what the publisher sells. Read any one of them against three questions (what is changing, who is struggling, what could help them), with word-for-word quotes and page numbers, the third answer marked as your inference, and every figure matched to its quote. Every add, drop and replace waits for the owner's yes. Fifteen read-only checks run on the folder.
Source Shelf: five rules and a five-check script (free sample)
From the Source Shelf: five rules for a small list of industry reports (one owner, link to the publisher not the post, dated and checked, three questions every time, every figure with its quote), and a read-only script that checks one small file for them.
Compounding Offer Audit: score an offer on six tests, find its weakest point, change one thing
Score each thing you sell from 0 to 3 on six tests: recurring, productized, systemized, owned demand, compounding, and whether growing it would make the owner's week better. Every score needs a checkable fact and its source, or the word unknown and the fact you are missing; no 3 on the owner's say-so, and no total. The audit names the weakest test, treats a low life score as a veto on growth, and proposes one change with a success test and a date. Every change waits for the owner's yes. Fifteen read-only checks run on the folder.
Research bundle: from first complaint to a decision you can defend, on one desk a script can check
Problem Mining Loop, Research Contract, Check the Claim and Compile Your Research in full, tied into one loop by a new Research Desk: every question names the decision it feeds, moves one stage at a time only when that stage's own checker exits 0, and draws on one shared register of sources with snapshots and claims with five verdicts. Someone who did no research audits before anything is compiled, fifteen read-only checks name every gap, and the owner's yes is typed at their own terminal and tied to the sha256 of the exact report.