An independent second opinion on your startup idea. Free, evidence-backed, ready in 10 minutes.
npx tweakideaWhat you get:
- A scorecard — verdict (PIVOT / GO / STOP), weighted score, letter-graded evidence
- A 14-dimension breakdown — per-dim scores with what's strong, what's weak, what's untested
- An assumption ledger — every untested claim in your idea, surfaced so you know what to validate next
- A shareable HTML report — opens in your browser, easy to revisit
Why not just ask ChatGPT?
- 14 evaluators run in parallel — one per dimension
- Each runs its own independent targeted web search and cites sources
- Every untested assumption in your idea is surfaced
- Output is a structured beautiful scorecard you can revisit and share, not a one-shot paragraph
Browse full interactive reports: Airlock · Mirror · Giftwrap
An idea dies from what you don't verify. TweakIdea runs an idea through 14 weighted dimensions — pain intensity, willingness to pay, solution gap, founder-market fit, and ten more — each scored with cited web research. Fourteen Sonnet evaluators work in parallel; Opus merges the results into a scorecard with strengths, weaknesses, and three rewrites targeting the weakest dimensions. The output names what you haven't tested yet. It doesn't tell you to quit your job.
Longer background: TweakIdea — the origin story · TweakIdea v2 — what changed
You: /tweak:evaluate "A mobile app that lets restaurants sell unsold
food at a discount 30 minutes before closing"
Claude: [researcher kicks off in background — web search on market,
competitors, pricing]
[extracts hypotheses from your idea — e.g. "restaurants have
unsold food at close", "30-min window is enough to sell it"]
Mark each hypothesis: CONFIRMED / UNCONFIRMED / REJECTED?
You: [confirms what you already know; leaves the rest unconfirmed]
Claude: [2-4 founder-fit questions on your connection to THIS idea —
restaurant ops? distribution? capital?]
You: [answers inline]
Claude: [launches 14 parallel evaluators, one per dimension]
[each runs its own targeted web searches for evidence]
[compute.py aggregates per-dim scores, weights them, grades
evidence A+ through F]
[Opus synthesizes strengths/weaknesses, next steps, and a
potential-uplift narrative driven by your unconfirmed
hypotheses]
PIVOT — Promising, address weak areas | Weighted 3.4/5.0
| Potential 4.0/5.0 | Evidence C+
[14-dim scorecard; report.html opens in browser]
You: /tweak:improve latest
Claude: [three rewrites — small reframe, medium reshape, big reimagine]
[each targets the weakest dimensions with rubric-grounded rationale]
[paste any rewrite into /tweak:evaluate to re-score]
| Weight | Dimension | What it measures |
|---|---|---|
| 12% | Pain Intensity | Painkiller vs. vitamin vs. candy |
| 12% | Willingness to Pay | Budget exists and buyer is reachable |
| 12% | Solution Gap | Why this hasn't been solved yet |
| 12% | Founder-Market Fit | Founder's domain, network, capabilities |
| 8% | Urgency | Forcing functions and active revenue loss |
| 8% | Frequency | How often the problem occurs |
| 8% | Market Size | TAM/SAM/SOM viability |
| 8% | Defensibility | Moats: network effects, switching costs, data |
| 4% | Market Growth | Sector CAGR trajectory |
| 4% | Scalability | Margins, self-serve, automation potential |
| 4% | Clarity of Target Customer | ICP specificity and findability |
| 4% | Behavior Change Required | Drop-in (5) vs. massive change (1) |
| 2% | Mandatory Nature | Regulatory or contractual forcing |
| 2% | Incumbent Indifference | Risk of being in the kill zone |
Runs 14 independent Sonnet evaluators in parallel — one per dimension — then merges them into a weighted scorecard with assumption tracking and letter-graded evidence. Produces a markdown scorecard and an HTML report.
/tweak:evaluate "A mobile app that lets restaurants sell unsold food at a discount"— inline/tweak:evaluate path/to/idea.md— from a file/tweak:evaluate— fully interactive
Searches Hacker News via Algolia and prints a ranked table of threads worth feeding into /tweak:analyze-hn-post. Each hit is scored 1–5 on how likely it is to yield a real startup-idea seed. Read-only.
/tweak:browse-hn llm agents week— topicllm agents, last 7 days/tweak:browse-hn robotics— topic only; prompts for a time window/tweak:browse-hn today— window only; browses everything recent/tweak:browse-hn— fully interactive
Fetches a Hacker News post (article + full comment tree), identifies 3-6 technology shifts, and surfaces 1-3 product opportunities per shift. Confirmed opportunities are written up and can be fed into /tweak:evaluate.
/tweak:analyze-hn-post 43374458— by HN id/tweak:analyze-hn-post https://news.ycombinator.com/item?id=43374458— by URL/tweak:analyze-hn-post— prompts for an id
Reads a completed evaluation run and generates three rewrites at escalating scales — Small reframe, Medium reshape, Big reimagine — each targeting the weakest dimensions with rubric-grounded rationale and mandatory trade-off acknowledgments. Read-only.
/tweak:improve latest— improve the most recent run/tweak:improve 20260416— improve a run by timestamp prefix
Compares two evaluation runs side-by-side — score changes, potential shifts, verdict movement, and a per-dimension breakdown of what moved. Read-only.
/tweak:diff latest— compare the two most recent runs/tweak:diff 20260413 20260416— compare two runs by timestamp prefix/tweak:diff 20260413 latest— compare a specific run against the latest
Lists artifacts under ~/.tweakidea/ — evaluation runs, HN analyses, and founder profiles. Arguments are free-form. Read-only.
/tweak:list— all categories, 5 most recent per section/tweak:list runs 20— the 20 most recent evaluation runs/tweak:list best ideas— runs ranked by weighted score/tweak:list hn— just HN analyses
Opens any artifact under ~/.tweakidea/ by timestamp, keyword, HN id, founder name, or natural query. Run reports open in the browser; HN and founder artifacts are inlined.
/tweak:show latest— most recent evaluation run/tweak:show 20260412-143022— a specific run by timestamp/tweak:show restaurant food waste— keyword match on idea text/tweak:show hn 43374458— an HN analysis by id/tweak:show best ideas— ranked query over saved runs
Uploads a run's report.html to a secret GitHub gist and prints two links — the gist itself and an htmlpreview.github.io rendered URL you can paste into Slack or email. Requires the GitHub CLI (gh auth login).
/tweak:share— share the most recent run/tweak:share latest— same, explicit/tweak:share 20260412-143022— a specific run by timestamp/tweak:share restaurant food waste— keyword match on idea text
Important
Requires access to both Claude Sonnet (14 evaluators + researcher + extractor) and Claude Opus (narrative synthesis). Without both, the pipeline won't complete.
- Claude Code installed
uv— needed by/tweak:browse-hnand/tweak:analyze-hn-postfor Python script execution. Install:curl -LsSf https://astral.sh/uv/install.sh | shorbrew install uv
npx tweakideaThe installer prompts for global (~/.claude) or local (./.claude) placement. After install, open Claude Code and type /tweak: — you should see browse-hn, diff, evaluate, improve, list, share, show, and analyze-hn-post in the autocomplete.
Note
First /tweak:evaluate run takes 10-20 minutes — it creates your founder profile, runs the researcher, and dispatches 14 parallel Sonnet evaluators plus one Opus synthesis pass. Subsequent runs reuse the founder profile and finish faster.
Uninstall
npx tweakidea -uBetter article extraction (optional)
For JS-heavy sites, run uv run playwright install chromium once. Without it, the HN fetcher falls back to plain HTTP — which works fine for most sites.
MIT — see LICENSE for details.