Default: k=1.0, α=1.0, β=0.20, sigmoid centre=0.10, slope=14. Tune to match your domain.
Files parse in your browser — text is extracted client-side and only the extracted content is sent to your chosen LLM provider when you click RUN. ~30k characters per document, ~100k total per request. The ✓ badge shows successful extraction.
Try: "What's my P_success?", "Run scenario X", "Compare to [benchmark]", "How sensitive is the result to [parameter]?"
P_success Q&A Platform · Beta · Effective immediately upon use
SynesisLabs.ai's P_success Q&A Platform (the “Platform”) is a research-led decision-support tool that scores projects on a 14-parameter probability-of-success model. The Platform is provided on a beta basis for evaluation and is not a financial-product, legal, or investment advisory service. Use it as a structured way to think about project risk; do not use it as a substitute for licensed professional advice.
The Platform runs entirely in your browser. To enable the AI-powered features (RUN MODEL, RUN scenarios, RUN benchmarks, executive summary), you must supply your own API key from a third-party large-language-model provider — currently Anthropic Claude or OpenAI. SynesisLabs.ai does not host, intermediate, or operate any backend that handles your data. When you click any RUN button, your browser sends the request directly to the provider you selected, over an encrypted (HTTPS / TLS) connection. No copy of your data passes through SynesisLabs.ai infrastructure.
Sent to your chosen LLM provider on each RUN action:
Stored in your browser only (localStorage):
Never sent to SynesisLabs.ai: any data you input or upload. We have no servers receiving your traffic. The Platform is a static-hosted single-page application.
By configuring an API key, you accept that the relationship for data processing is between you and the LLM provider you selected, governed by their commercial terms.
It is your responsibility to read, understand, and accept those terms before clicking any RUN button. If a provider's terms are not acceptable to your organisation, do not configure a key.
SynesisLabs.ai treats inputs to the Platform as your confidential material — except that we never receive them. The chain of custody is yours and your provider's. We recommend you do not upload material classified as defence-controlled, regulated personal health information, restricted personal data, or any data subject to contractual confidentiality with a third party, unless you have first confirmed your provider's terms permit it. The default Anthropic and OpenAI commercial offerings are usually acceptable for ordinary business material under NDA; for stricter requirements (defence, healthcare, regulated personal data) you should pursue Zero Data Retention or a backend-proxy architecture before using this Platform on that material.
SynesisLabs.ai is responsible for the Platform code: how it parses documents, how it constructs prompts, how it stores credentials in localStorage, how it presents results, the system prompts used, and the third-party libraries it loads (PDF.js, Mammoth, SheetJS, PapaParse, and the LLM-provider SDKs). We endeavour to follow reasonable security practices but provide the Platform on an as-is basis during the beta. If you discover a security defect, please report it.
You are responsible for: your relationship with your chosen LLM provider; the accuracy and confidentiality status of any data you paste, type, or upload into the Platform; safekeeping of your own exported project files; obtaining any internal approvals, NDAs, regulatory permissions, or export-control reviews required for the material you input; and the conclusions you draw from the P_success score and any AI-generated rationales.
P_success values, executive summaries, scenario impacts, benchmark comparisons, and Monte Carlo distributions produced by the Platform are model outputs. They are not financial-product advice (Australian Corporations Act 2001 (Cth) s766B; s911A do not apply because no advice is provided), not legal advice, not investment advice, and not a substitute for due diligence. Outputs reflect the parameter values you supply or that the LLM proposes from your inputs; small input changes can produce materially different outputs. Treat the Platform as an analytical aid, not a forecast.
The Platform may reference third-party companies and public information (e.g., named benchmarks like Stripe, Moderna, IBM Quantum) for illustrative comparison. SynesisLabs.ai makes no claim of endorsement by, or affiliation with, those parties. All trademarks remain the property of their respective owners. Materials you upload remain your property and your responsibility; ensure you have the right to share them with your selected LLM provider.
To the maximum extent permitted by law, the Platform is provided "as is" and "as available", without warranty of any kind, express, implied or statutory. SynesisLabs.ai expressly disclaims all warranties including (without limitation) warranties of merchantability, fitness for a particular purpose, accuracy, completeness, reliability, security, non-infringement, uninterrupted operation, error-free operation, or that defects will be corrected. SynesisLabs.ai makes no representation that P_success scores, executive summaries, scenarios, benchmarks, Monte Carlo distributions, or any other output of the Platform are accurate, current, complete, suitable for any particular purpose, or fit to be relied upon for any decision.
To the maximum extent permitted by applicable law, in no event shall SynesisLabs.ai, its directors, officers, employees, contractors, agents, partners, licensors or affiliates be liable to you or any third party for any of the following arising out of or in connection with your use of, inability to use, or reliance on the Platform:
Aggregate cap. SynesisLabs.ai's total aggregate liability for all claims arising out of or relating to the Platform, whether in contract, tort (including negligence), strict liability or otherwise, shall not exceed the greater of (a) fees paid by you to SynesisLabs.ai for the Platform in the twelve (12) months preceding the event giving rise to the claim, or (b) AUD $100. During beta, fees paid are zero, so the AUD $100 floor applies.
Carve-outs. The exclusions and limitations above do not apply to (i) liability for fraud or fraudulent misrepresentation, (ii) liability for death or personal injury caused by SynesisLabs.ai's negligence, or (iii) any other liability that cannot be excluded or limited under applicable law (including, without limitation, certain rights under the Australian Consumer Law). Nothing in these terms is intended to exclude or limit any non-excludable consumer guarantees.
You agree to defend, indemnify and hold harmless SynesisLabs.ai, its directors, officers, employees, contractors, agents, partners, licensors and affiliates from and against any and all claims, demands, suits, proceedings, losses, liabilities, damages, judgments, awards, settlements, costs and expenses (including reasonable legal fees) arising out of or relating to:
SynesisLabs.ai shall not be liable for any failure or delay in performance arising from causes beyond its reasonable control, including but not limited to: acts of God; natural disasters; pandemics or epidemics; war, terrorism, riots or civil unrest; government actions, sanctions, or export controls; outages or disruptions of the internet, undersea cables, electricity grids or telecommunications networks; failures of LLM providers, cloud service providers, hosting providers or content delivery networks; cyber-attacks; or any other event of similar gravity.
You agree not to use the Platform to: (a) violate any applicable law or regulation; (b) infringe the intellectual property, privacy, publicity, confidentiality, contract or other rights of any person; (c) generate, evaluate, or facilitate the development of weapons (chemical, biological, radiological, nuclear, or conventional); (d) generate or evaluate content that exploits or endangers minors or vulnerable persons; (e) deceive or defraud any person, including by submitting falsified material to the LLM as if it were authentic; (f) attempt to reverse-engineer, decompile, disassemble, scrape, or extract the Platform's proprietary methodology, equations, calibration constants, or system prompts for the purpose of building a competing product; (g) circumvent any technical or contractual restriction; or (h) cause harm to SynesisLabs.ai, its users, its providers, or any third party. SynesisLabs.ai may suspend or terminate your access to the Platform without notice if you breach this section.
You acknowledge and agree that: (i) you are not relying on any representation, warranty, or statement made by SynesisLabs.ai except as expressly set out in these terms; (ii) the Platform's outputs are model-generated estimates based on your inputs and on third-party LLM reasoning, and may contain errors, omissions, hallucinations, or biases; (iii) you have made your own independent assessment of the Platform's suitability for your purpose; (iv) you assume all risk arising from any decision, action, transaction or omission you take in reliance on Platform outputs; and (v) before making any material decision, you will conduct independent verification and obtain advice from suitably qualified professionals.
Any claim or cause of action arising out of or relating to the Platform must be commenced within twelve (12) months after the cause of action accrues. After this period, such claim or cause of action is permanently barred, to the maximum extent permitted by applicable law.
These terms and any non-contractual obligations arising out of or in connection with them are governed by the laws of the State of Victoria, Australia, without regard to its conflict-of-laws principles. The parties submit to the exclusive jurisdiction of the courts of Victoria, Australia, and the federal courts of Australia situated therein. (Customise this section for your jurisdiction before a commercial release.)
The Platform's source code, including the P_success equations, calibration constants, system prompts, parameter taxonomy, and report templates, is the confidential intellectual property of SynesisLabs.ai. Beta users are granted a limited, revocable, non-exclusive, non-transferable licence to use the Platform for internal evaluation only. Any redistribution, sublicensing, reverse engineering for competitive product development, or republication of methodology requires written permission. This obligation survives termination.
SynesisLabs.ai may terminate or suspend your access to the Platform at any time, with or without cause and with or without notice. Upon termination, your right to use the Platform ceases immediately. Sections 10 (No warranty), 11 (Limitation of liability), 12 (Indemnification), 14 (Acceptable use, in respect of past breaches), 15 (Non-reliance), 16 (Time limit on claims), 17 (Governing law), 18 (Confidentiality), and any other provision that by its nature should survive, will survive termination.
If any provision of these terms is held to be unenforceable, that provision will be enforced to the maximum extent permissible and the remaining provisions will remain in full force and effect. These terms, together with the LLM provider's terms you accept separately, constitute the entire agreement between you and SynesisLabs.ai regarding the Platform and supersede all prior or contemporaneous communications. No waiver of any provision shall constitute a continuing waiver. Headings are for convenience only and do not affect interpretation.
These terms may be updated as the Platform evolves out of beta. Material changes will be announced via the in-app notice. By continuing to use the Platform after a change is announced, you accept the updated terms. Questions: [email protected].
This document is a beta-stage user notice and not a substitute for a formal Terms of Service or Privacy Policy. SynesisLabs.ai recommends qualified legal review before commercial release. Last updated: .
Treat each parameter as a probability distribution rather than a fixed value. Sample around your current baseline N times and compute P_success for each draw — turning a single point estimate into a confidence band.
Each parameter is perturbed by ±sigma × its full range, drawn uniformly. Higher uncertainty → wider P_success band.
This is a research-preview / beta platform. Please read before using with confidential or third-party material. Last updated .
SynesisLabs.ai's P_success Q&A Platform (the "Platform") is a research-led modelling tool that turns project information into a probability-of-success score, scenario stress tests, and a structured report. Outputs are decision support, not financial, legal, medical or other regulated advice.
The Platform is provided "as is" for evaluation purposes. The underlying P_success algorithm, calibration constants, system prompts, and parsing pipeline are proprietary to SynesisLabs.ai. Reverse engineering, redistribution, or commercial use without written permission is not authorised.
The Platform runs entirely in your browser. SynesisLabs.ai does not operate any server that receives, processes, or stores your project data. When you provide an API key for an LLM provider (Anthropic, OpenAI, etc.), that key is held in your browser's local storage and used to call the provider directly from your machine.
This means you have a direct relationship with your chosen LLM provider. You are responsible for accepting and complying with their commercial terms (e.g. Anthropic's commercial terms, OpenAI's business terms), their usage policies, and any data-handling commitments they make to you.
Stays in your browser only: uploaded reference documents (raw files), the project description, slider values, scenario shifts, benchmark scores, calibration constants, the API key.
Sent to your chosen LLM provider over HTTPS when you click RUN MODEL, RUN scenarios, RUN benchmarks, or Test connection: the project description, the project name, the extracted text content of any uploaded documents (capped at ~30,000 characters per document and ~100,000 characters per request), and the LLM provider's response back to you. With web search enabled, the LLM provider may additionally route formulated search queries through their search backend.
Sent to SynesisLabs.ai: nothing. We have no server collecting your data.
Persisted across sessions in your browser's localStorage: the API key (until you click Clear), the API verification flag, and (when you Export) full project state including parsed document text. Treat exported project JSON files as sensitive — they contain the same content the LLM saw.
SynesisLabs.ai cannot retain your data, but the LLM provider you select may. Most commercial LLM APIs retain request and response logs for a period (Anthropic: approximately 30 days for trust & safety review; not used for model training). For stricter retention requirements, your provider may offer enterprise options such as Zero Data Retention agreements, regional data residency, or processing via cloud marketplaces (AWS Bedrock, Google Vertex AI). You should consult your provider's documentation directly.
When web search is enabled in the API panel, your selected LLM provider may, at its discretion, fetch live web pages to ground its reasoning. Search queries are formulated by the model from the context you supplied. Page content fetched in the process is processed by the LLM provider; SynesisLabs.ai has no role in that step. You can disable web search at any time in the API panel.
The Platform's outputs (P_success scores, scenarios, benchmarks, executive summaries, reports, Monte Carlo statistics) are decision-support artefacts only. They are not financial product advice, investment advice, legal advice, medical advice, regulatory advice, or any other regulated service. SynesisLabs.ai does not hold an Australian Financial Services Licence (AFSL) and does not purport to provide such services. You should consult appropriately licensed professionals before making decisions of consequence.
The Platform is provided to you for evaluation under confidentiality. The source code visible in your browser, the system prompts, calibration constants, parsing pipeline, scenario libraries, and benchmark archetypes are SynesisLabs.ai's confidential information. You agree not to reproduce, redistribute, or share access without written permission.
The Platform is provided on an "as is" basis without warranties of any kind, express or implied. SynesisLabs.ai disclaims all implied warranties of merchantability, fitness for a particular purpose, accuracy, completeness, and non-infringement to the maximum extent permitted by law.
To the maximum extent permitted by law, SynesisLabs.ai's aggregate liability for any claim arising from your use of the Platform shall not exceed the amount you paid for the Platform in the 12 months preceding the claim (which, during the beta period, is zero). SynesisLabs.ai shall not be liable for indirect, consequential, special, incidental, or punitive damages.
The Platform is in active development. Features may change, break, or be removed without notice. You should re-read these terms periodically; the "Last updated" date reflects the current version.
Questions about these terms or the Platform: contact SynesisLabs.ai through the channel by which you received access.
© SynesisLabs.ai. All rights reserved. This document is supplied as part of the Platform; the version on this page governs in case of conflict with any out-of-band copy.
Wire research, scenario, and benchmark proposals to a real LLM. Your key is retained across sessions in this browser's local storage — close the tab and come back, the key is still loaded. Use Clear key to remove it entirely.
Status indicator: grey dot = no key. Solid green = key loaded but not yet tested. Blinking green = last Test connection succeeded.
Deterministic mode is on: calls use temperature 0 (and a fixed seed for OpenAI) so the same project description should produce the same parameter values across runs.
Web search (toggle below) lets the model fetch live web pages — current valuations, recent news, named-competitor research — and cite sources. Strongly recommended; defeats the LLM's training-data cutoff.