Work
Research that drops bad sources and claim-checks every line
Confident AI research briefs and correct ones look the same — until a CEO blog claim goes viral.
What hurt
@Amir_Radarit tip: researching "does a 4-day work week work?" the useful move was throwing sources away. Score sources — drop anonymous / no-methodology; keep peer-reviewed, raw data, primary, large-n. Then claim-check every line: SUPPORTED / WEAK / UNSUPPORTED with independent-source counts. A brief with three UNSUPPORTED labels beats a brief where everything sounds certain. Tip only (setup DM gated; no invented prompt). Not affiliated with xAI.
What to install or create
Source map → drop bad sources → keep quality tags → claim-check SUPPORTED/WEAK/UNSUPPORTED.
Bots
No new bot. Follow the first-run steps.
Plugins
Browser
Optional
Routines
None on the first run. Watch one manual pass first.
The prompt
This is a tip, not a pasteable system prompt. The author did not publish one. Follow the first-run steps.
First run
- 1Read https://x.com/Amir_Radarit/status/2117750968326529515 (credit @Amir_Radarit).
- 2Tell your research bot: drop anonymous / no-methodology sources; tag keepers (peer-reviewed, raw data, primary, n=…).
- 3Require a claim table: each claim → SUPPORTED / WEAK / UNSUPPORTED + independent-source count.
- 4Ask for the dropped-source list next to the kept ones — red X reasons visible.
- 5Approve the brief only after you scan UNSUPPORTED/WEAK rows; send or publish nothing yet.
Approval fence
Show the source map and claim table first — never publish or email the brief without your yes.
What to tweak
Steal drop-rules + claim verdicts; dropped sources matter as much as kept ones.
Source
Amir @Amir_Radarit · Original post
Not affiliated with xAI. We did not write this prompt. Third-party template terms still apply.
Pro pack
Want the full setup, not just the idea?
Pro packs give you the job text, prompts, schedules and pitfalls for popular use-cases.