Research bundle: from first complaint to a decision you can defend, on one desk a script can check
by PaperChaseWebb
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.
- Price
- $59
- Difficulty
- Intermediate
- Setup
- 1 hr
- Format
- .md file
research-bundle-from-first-complaint-to-a-decision-on-one-desk.md · 338 KB · instant download after checkout · $101 if bought separately
Want this built for you instead? We can build it for you.
What you end up with
Every open research question in one place, with the stage it is at, the checks it passed, who audited it and the exact report you said yes to, so you know which answers are finished before you act on them.
What you get
- Four packs in full: Problem Mining Loop, Research Contract, Check the Claim and Compile Your Research ($101 bought separately)
- The Research Desk (only in this bundle): one folder of plain files for questions, handoffs, sources, claims, audits and approvals, with templates
- The research loop: which pack each stage uses, what it hands on and which checker must pass
- desk.py: handoffs that run each stage's checker and record only on exit 0, owner approval at a terminal tied to the report sha256, verify with expiry
- desk_check.py: fifteen read-only checks, plus claim_cards.py for the claim stage, with a seeded test desk (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 nine research packs and research skills, with sources
Who it is for
Founders, consultants and small teams who use agents to research before a real decision and want to know which answers are actually finished.
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.
What is in the bundle (4 packs)
| Pack | Format | Separate price |
|---|---|---|
| Problem Mining Loop: real complaints, graded and counted, before you build anything | Prompt pack | $19 |
| Research Contract for Agents: claims, sources and dates your agent must keep | Prompt pack | $19 |
| Check the Claim (expanded edition): a method, prompts and a skill for checking AI claims | Prompt pack | $39 |
| Compile Your Research: a cited, checked second brain your agents keep | Prompt pack | $24 |
| Bought separately | $101 | |
| The bundle | $59 (you save $42) |
Tools needed
Cost to run: The bundle is the only purchase. The checkers and desk.py use no model tokens; the research itself uses whatever your agent normally spends.
Time to result: A ratified desk with your open questions on it in one afternoon once the four packs are installed
How it works
Run the tests
Copy the research-bundle folder and run sh test_all.sh.
Expected: 15 findings on the seeded desk and 0 on the fixed one, checker tests 35 passed, desk tests 25 passed, exit 2 on a missing folder.
Install the four packs
Set up Problem Mining Loop, Research Contract, Check the Claim and Compile Your Research, and run each one's own tests.
Expected: Each pack's checker runs from your machine.
Start the desk
Run desk.py init research/desk, fill desk.json with your name, an auditor, your researcher agents and a review date, then ratify it with desk.py approve --desk.
Expected: desk_check.py shows no R01, R02 or R15 findings.
Load your questions
Put each open question in questions.csv with the decision it feeds, who asked and a due date, then run desk.py status.
Expected: Each question shows its next step and the pack it uses.
Work, audit, decide
Hand each question on with desk.py handoff, get the audit before compile, then approve the report yourself and run desk.py verify before you act.
Expected: Decided questions verify as approved, and any edit to a report shows up at once.
Sample prompt from the pack
Run desk_check.py on research/desk. Show me the findings in order, the smallest fix for each and the desk hash. Edit nothing until I say yes.
Prompt 3 of 10.
For Q-id, put every claim from the brief in claims.csv with one of the five verdicts. Save each source once to sources.csv with a snapshot. Run claim_cards.py and fix what it finds before handing off.
Prompt 7 of 10.
Free preview of the file
# Research bundle: from first complaint to a decision you can defend, on one desk a script can check Version 1.0, 2026-10-29. By PaperChaseWebb. A bundle of four paid packs (Problem Mining Loop, Research Contract, Check the Claim and Compile Your Research) plus a new Research Desk that ties them into one research loop. The desk comes with a handoff and approval script, a read-only checker with fifteen checks, a claim-stage checker, templates, a seeded test desk, 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 10: everything ran on Linux with Python 3.13, including owner approvals typed at a pseudo-terminal. The skill file, snippet and rules block were not loaded inside the harnesses for this release, and the desk has not been run on a real research question. Each limit is marked where it matters. Most research packs for agents are a long prompt and a promise. "Deep research in 60 seconds." "200 prompts." The better open-source research skills go further, with source tiers, citation checks and a critic pass, and they are worth reading. What none of them hold is the part between one report and the next. A complaint someone found last month never became a proper question. A brief was written but the claims inside it were never checked. A claim was checked, then quietly reused in a later report after the page b
Test record
- Date
- Oct 29, 2026
- What was run
- desk_check.py on a seeded desk 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 desk, 0 on the fixed one; 15 of 15 planted faults caught alone; exit 2 on missing input; nothing written. 35 checker and 25 desk.py tests passed, 0 failed.
- Not run
- The agent files inside Claude Code, Codex, Hermes or Grok Bot. A real research question. Windows and macOS.
Copied from this item's dated test notes. What was not run stays in.
Worth knowing
- The checker checks that each research step was done and written down. It cannot tell whether a source really says what a claim says; the auditor and the included packs' own checks do that.
- 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, and the desk has not been run on a real research question. The test desk is a made-up pottery studio. Linux only; Windows and macOS were not tested.
- The approval script 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 3.13.5: desk_check.py gave 15 findings on the seeded desk (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 file or broken desk.json; checker tests 35 passed, 0 failed; desk.py tests at a pseudo-terminal 25 passed, 0 failed.
- The bundle contains the versions listed on this page on the day you buy. Each included pack keeps its own notes and limits.
Questions
Why not buy the four packs on their own?
You can, for $101. The bundle is $59 and adds the Research Desk, which is only here: one place for every question, handoffs that need each checker to pass, one source and claim register, audits and your yes tied to the exact report.
Does it do the research for me?
No. Your agent does the research with the four packs. The desk keeps track of where each question is and refuses to call it finished before the checks, the audit and your yes.
Can my agent approve a report?
No. desk.py approve refuses piped input and the agent environments it recognizes. You type your name and part of the report hash at your own terminal.
What if a report changes after I approve it?
Its sha256 changes, desk_check.py flags it and desk.py verify fails until you read it and approve it again.
What if I want it set up with me?
Book a call on the paperchasewebb.ai homepage about the $5,000 Install sprint, or Manage at $5,000 a month to keep the desk and morning checks running.
Details
Version 1.0 · Updated Oct 29, 2026 · Markdown version of this page
License: Use on your own machines and for research you do for clients and teams. Do not resell or republish the bundle itself. Download terms
Related
Research Desk: five rules for research you will act on, and a five-check script (free sample)
From the Research bundle: five rules for research that feeds a decision (an owner and an auditor who did not do the research, a decision named for every question, one of five verdicts per claim, a date on every source, a primary source behind every CONFIRMED), and a read-only script that checks one small file for them.
Problem Mining Loop: real complaints, graded and counted, before you build anything
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.
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.
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.