Paid laboratory instruments
Inspect the evidence. Then decide whether to buy.
YY's paid products answer narrow operational questions. Each product exposes its buyer, supported environment, public output, price, and non-claims before checkout.
- Public sample first
- Exact supported boundary
- No subscription required
Self-serve software
Two local, one-time licenses
Both releases run on customer-controlled machines and are delivered through Polar. Neither includes an SLA or promises that a future feature will ship. Any 0.x update that is released is included as stated on the product page.
- Founding release · v0.2.0
Dataset Preflight
Inspect a local dataset for duplicates, provenance gaps, and secret-risk signals, enforce your thresholds, and emit opt-in SARIF and JUnit evidence for CI.
Best fit: Teams reviewing a local dataset directory before upload, handoff, or a training pipeline run.
Boundary: Not a security audit, legal or license clearance, semantic duplicate detector, or training-readiness verdict.
- $39 once
- Python 3.12
- No runtime network requests
- Five-artifact public sample
- Founding release candidate · 0.0.1-rc7
AgentSafe
A deterministic, local evidence checker for one specific behavior: approval-to-execution in OpenAI Agents SDK 0.19.x FunctionTools.
Best fit: OpenAI Agents SDK teams using instrumented FunctionTools with static approval policies.
Boundary: Not a sandbox, penetration test, security audit, certification, or proof that an agent is safe.
- $49 once
- Python 3.12
- OpenAI Agents SDK 0.19.x
- Six-check public sample
Third-party metering
One paid Poe utility
Chat Receipt is metered by Poe rather than sold as a YY software license. Poe controls its base point charge and checkout surface.
- Paid per message · Poe
Chat Receipt
Export the conversation visible to the bot as a deterministic ZIP with a readable transcript, canonical JSON receipt, SHA-256 manifest, and local verifier.
Exports only the history Poe presents to the bot. Verification proves packet consistency, not truth, identity, origin, or completeness.
- 501 Poe points per message at publication
- No downstream model call
- Deterministic ZIP + local verifier
Before checkout
Buy only inside the measured boundary
- 01
Open the sample
Inspect the output shape and the kinds of uncertainty the product preserves.
- 02
Check your environment
Confirm the exact Python, framework, tool, and operating assumptions on the product page.
- 03
Keep the conclusion narrow
Use the result as evidence for a reviewer, never as a certification or broad safety verdict.