Launch App

Your prompt is the ceilingon every answer you get.

Researchers have published hundreds of thousands of prompting techniques. You know maybe five. Our source reads that flood at a scale no person can and keeps what earns keeping; we index that part, 6,235, every one lifted from a published paper you can open and check.

Then we throw almost all of them away. For your task: 150 close candidates, 10 read in full by separate AI specialists, 3 ranked against each other by a judge that shows its reasoning. You pick the winner and it lands as a prompt you can paste. One method out of six thousand, chosen for the job in front of you.

Free to join, no card · your first full run is on us · we checked 1,420 of these effects against the papers that reported them.

SELECTION FUNNELRUNNING
Techniques indexed6,235
Semantic candidates150
Read in full by specialists10
Ranked by the judge3
Prompt generated1
7 stages traced per run

MEASURED PERFORMANCE

ValueParameterMethod
6,235Techniques indexedextracted from published AI research, one source document each
98.5%Retrieval recallmeasured against our ground-truth benchmark
10Specialist agentsone agent per candidate, run in parallel on every search

The advice "just write better prompts" is useless.

You've seen it happen. The same model that writes brilliant analysis for a colleague gives you a wall of generic filler.

The difference is not the model. It's the technique. Hundreds of papers have measured which prompting patterns work for which kind of task.

Nobody has time to read them. So most people recycle the same three habits and accept whatever comes back.

Note

Every shallow answer costs you twice: once when you read it, and again when you make a decision based on it.

What happens in the ninety seconds after you hit search

01input

You describe the task.

Plain language, any of 97 languages. Attach up to 10 files: PDF, Word, Excel, images.

≤ 10 filesaccepted per task
02search

It searches the way a research librarian would.

Your intent gets translated into the vocabulary of the research literature, then matched against 6,235 techniques. Near-duplicates get pushed apart, so the shortlist holds ten different ideas instead of the same idea ten times.

6,235 → 150candidates after semantic pass
03analysis

Ten specialists read the actual research.

Each finalist gets its own agent: one technique, the full source document behind it, about 5,000 words. No skimming, no summary-of-a-summary.

~5,000 wordsread in full, per candidate
04verdict

A judge picks the top three and shows its work.

A reasoning model with a 15,000-token thinking budget compares all ten reports and ranks the top three: strengths, weaknesses, when to avoid. You pick. It never picks for you.

15,000 tokensjudge thinking budget
05output

You get a prompt built for your task.

A complete prompt that applies the chosen technique to your task, generated on whichever of 7 models you choose. No blanks to fill in.

7 modelsbudget to premium tiers

This is what a decision looks like

Real screens from the live app. No mockups.

PLATE 1app.getdecision.ai · verdict
GetDecision verdict view: three techniques ranked with fit scores, strengths and considerations

Fig.Three techniques ranked by fit score, with strengths and considerations. You pick, it never picks for you.

PLATE 2describe the task
GetDecision input step: describe your task in plain language

Plain language in, any of 97 languages. Files welcome.

PLATE 3search running · live
GetDecision processing step: live pipeline progress with per-stage timing

Every stage streamed live while the ten specialists and the judge work.

Watch your search think

Every search produces a full trace: seven pipeline stages, per-stage inputs and outputs, and token counts. A live diagram shows candidates flowing from 6,235 down to your top 3.

If a recommendation looks off, open the trace and see where the reasoning went.

trace · session_8f3a✓ complete
01 input analysis0.4s
02 intent reformulation1.1s
03 vector search · 6,235 → 1500.6s
04 document scoring · 150 → 105.9s
05 specialists ×10 · parallel4.8s
06 judge · deep reasoning75.2s
07 prompt generation8.0s
total95.8s

Everything around the pipeline works today

Session history

Every search, choice, and generated prompt saved and searchable.

Favorites and collections

Organize techniques by project. Drag and drop.

File-aware prompting

Your uploaded documents shape both the search and the final prompt.

Model choice

7 models across budget, standard, and premium tiers.

Regenerate with another technique

Rebuild your prompt with any other technique from your Top-3 and compare side by side.

Simple monthly plans

Free
Try the full pipeline
$0
1 free run on signup
  • One full run: search + generated prompt
  • All 7 generation models
  • Full history & collections
  • No card required
Starter
For occasional decisions
$10/mo
1,000 credits / month
  • ≈ 30 full runs on premium models
  • ≈ 40 runs with Claude Sonnet 5
  • Credits never expire
  • Same features as every plan
Recommended
+20% credits
Pro
For weekly heavy use
$25/mo
3,000 credits / month
  • ≈ 90 full runs on premium models
  • ≈ 120 runs with Claude Sonnet 5
  • Credits never expire
  • Same features as every plan
+33% credits
Max
For teams & power users
$60/mo
8,000 credits / month
  • ≈ 240 full runs on premium models
  • ≈ 320 runs with Claude Sonnet 5
  • Credits never expire
  • Same features as every plan

A full run = one search (20 credits) plus one prompt generation (1–12 credits depending on the model).

Fair questions

Is this a prompt library?

No. Libraries give you 10,000 prompts and leave the choosing to you, which is the hard part. GetDecision does the choosing, with evidence, and then writes the prompt.

Why does a search take about a minute and a half?

Ten agents are reading full research documents and a judge is comparing their reports. You watch every stage live while it runs.

Which models does it use?

Gemini 2.5 Flash-Lite runs scoring and the ten specialists, Claude Sonnet 4.5 with extended thinking is the judge. For the final prompt you pick from 7 models: GPT-5.6 Sol, Claude Opus 4.8, Claude Sonnet 5, Gemini 3.1 Pro, Grok 4.5, and two budget options.

What does it cost?

Your first run is free, the search and the prompt both. After that, plans start at $10 a month. A search costs 20 credits, generating the prompt adds 1–12 depending on the model. Cancel anytime.

Where do the 6,235 techniques come from?

From published AI research. Each technique keeps its source reference, and the full extracted document travels with it through the pipeline.

The next hard question you ask deserves better than a guess.

Describe your task. Watch 6,235 techniques become 3. Take the prompt.

Launch App