About AESOP

• Start doing the AESOP things!

Non-Disclosure Agreement

AESOP Beta Testing AgreementThis Beta Testing Agreement (“Agreement”) is made between you (“Beta Tester” or “you”) and AESOP (“AESOP,” “we,” “us,” or “our”) regarding your participation in the beta testing of the AESOP platform. By accessing or using the AESOP beta platform, you agree to be bound by the terms of this Agreement, our Privacy Policy, and our Terms of Service, which are incorporated herein by reference.1. Purpose and Scope
AESOP invites you to participate in the beta testing of its platform designed for evaluating and analyzing AI models and agent behaviors. The primary purpose of this beta test is to collect feedback, identify bugs, and enhance the platform prior to its official release. Your participation is voluntary and governed by the terms outlined herein.2. Acceptance of Terms
By accessing, downloading, or using the AESOP beta platform, you acknowledge that you have read, understood, and agree to be bound by this Agreement, the Privacy Policy, and the Terms of Service. If you do not agree to these terms, you may not access or use the platform.3. Confidentiality and Non-Disclosure
You recognize that the AESOP beta platform, including all software, documentation, methodologies, algorithms, and any feedback or data generated during the beta testing phase, may contain confidential and proprietary information (“Confidential Information”). You agree to:
- Not disclose Confidential Information to any third party without obtaining prior written consent from AESOP.
- Use Confidential Information solely for the purpose of participating in the beta test.
- Take reasonable measures to safeguard Confidential Information from unauthorized access or disclosure.You agree to maintain the confidentiality of Confidential Information for a period of two (2) years following the termination of this Agreement. Upon termination of this Agreement, you agree to either return or destroy all copies of the Confidential Information in your possession and provide AESOP with written confirmation of such actions. The confidentiality obligations do not apply to information that:
- Is or becomes publicly known through no breach of this Agreement by you.
- Is independently developed by you without the use of or reference to the Confidential Information.
- Is required to be disclosed by law or regulation, provided that you give AESOP prompt notice of such requirement and cooperate with AESOP in seeking any protective order.You acknowledge that a breach of this Agreement could cause irreparable harm to AESOP, and therefore, in addition to any other rights or remedies that may be available at law or in equity, AESOP shall be entitled to seek injunctive relief to enforce your obligations under this Agreement.4. License Grant
AESOP grants you a limited, revocable, non-exclusive, non-transferable, non-assignable license to access and use the AESOP beta platform strictly for beta testing purposes. You may not:
- Modify, reverse engineer, decompile, or disassemble the platform or any portion thereof.
- Redistribute, sublicense, or commercialize the platform or any part of it.
- Use the platform for any purpose other than beta testing as directed by AESOP.5. Feedback and Intellectual Property
You agree that all feedback, suggestions, improvements, and contributions you provide (“Feedback”) are the sole property of AESOP. You hereby assign to AESOP all rights, title, and interest in and to such Feedback; AESOP may use such Feedback for any purpose without limitation or compensation.All content, layout, design, data, databases, and graphics on the platform are protected by intellectual property laws and are owned by AESOP or its licensors unless otherwise stated.6. Beta Tester Obligations
In consideration of your access to the AESOP beta platform, you agree to provide the following within two (2) weeks of gaining access:
- Testimonial: A short statement regarding your experience with the beta platform, which may be used by AESOP for promotional or marketing purposes.
- Focused Feedback: Detailed feedback on your experience using the beta platform, including usability, performance, feature suggestions, and any bugs encountered.You acknowledge that your testimonial and feedback will contribute to the improvement of the AESOP platform and may be incorporated into AESOP’s marketing materials. AESOP may contact you for follow-up information related to your feedback.7. Data Protection and Privacy
AESOP implements industry-standard security measures to protect your data and ensure compliance with applicable privacy regulations. All data transmissions and storage are encrypted using state-of-the-art encryption protocols.Information Collected:
- Account Information: Your email address, name, and authentication credentials.
- Usage Data: Information on how you use the platform, including evaluation sessions and agent interactions.
- Model Data: Data submitted for evaluation, processed strictly for evaluation purposes.How We Use Your Information:
- To provide and maintain evaluation services.
- To process and analyze AI model evaluations.
- To improve our platform and develop new features.
- To communicate with you about your account and our services.
- To ensure platform security and prevent misuse.
- To comply with legal obligations.Data Retention and Deletion:
Your personal information is retained only as long as necessary. You may request the deletion of your data at any time. Evaluation data is anonymized and aggregated for analytical purposes.Your Rights:
You have the right to access, update, or delete your personal information and to request a copy of your data or restrict its processing.8. Limitation of Liability and DisclaimerAESOP, its developers, contributors, and affiliated parties will not be liable for any direct, indirect, incidental, special, or consequential damages resulting from the use or inability to use the platform. The AESOP tool and associated documentation are provided “as is” without any express or implied warranties. You are solely responsible for verifying the accuracy, quality, and suitability of the evaluation results provided by AESOP. AESOP does not guarantee that the tool will function without errors or that the evaluation results will satisfy specific requirements.9. Indemnification
You agree to indemnify and hold harmless AESOP, its developers, contributors, and affiliated parties from any claims, damages, or losses arising from your use of the platform or your violation of this Agreement.10. Use Restrictions
You agree to use the AESOP platform only for lawful purposes and in a manner that does not restrict or inhibit anyone else’s use of the platform. You may not use the platform to:
- Transmit harmful, illegal, or offensive content.
- Attempt to gain unauthorized access to any part of the platform or its systems.
- Interfere with or disrupt the platform or its servers.11. Termination
AESOP reserves the right to terminate your access to the beta platform at any time, for any reason, without prior notice. Upon termination, you must cease all use of the platform and delete or destroy any Confidential Information in your possession.AESOP may monitor your compliance with this Agreement and reserves the right to take appropriate action, including termination of access and legal recourse, if any terms of this Agreement are breached.12. Governing Law
This Agreement shall be governed by and construed in accordance with the laws of the state of Colorado, without regard to its conflict of law provisions.13. Assignment and Notices
You may not assign or transfer your rights or obligations under this Agreement without the prior written consent of AESOP. Notices to you may be sent to the email address you provided during registration.14. Entire Agreement and Amendments
This Agreement, along with the Privacy Policy and Terms of Service, constitutes the entire agreement between you and AESOP regarding the beta testing of the platform. AESOP may amend this Agreement at any time by posting the revised version on the platform or notifying you via email. Your continued use of the platform after such amendments constitutes acceptance of the revised terms.---Contact Information
If you have any questions about this Agreement, the Privacy Policy, or the Terms of Service, or wish to exercise your rights, please contact us at: [email protected]

• Back to Beta Access

AESOP - Terms of Service

AESOP - Privacy Policy

Contact Us

AESOP - Ideation

The AI Ideation Agent is a specialized system designed to guide creators through the early conceptualization and development of AI systems using a structured, foundation-first interview approach. Unlike traditional requirement-gathering tools, this agent begins with core purpose and intention, ensuring that all subsequent technical and implementation decisions flow from a solid understanding of fundamental goals and desired impact.The agent operates through a progressive, layered interview process that starts with essential questions about purpose, stakeholders, and ethical considerations before moving into more detailed technical and implementation discussions. It employs an adaptive questioning framework that adjusts its depth and focus based on user engagement and project needs, while maintaining strict requirements for establishing foundational elements before proceeding to dependent topics. This ensures that critical aspects like ethical considerations and stakeholder impacts are addressed early, preventing the common pitfall of jumping to technical solutions before fully understanding the problem space.Throughout the interview process, the agent maintains comprehensive real-time documentation, creating relationship maps between decisions and tracking dependencies. It uses a sophisticated evaluation framework that weighs different aspects of the project to ensure balanced consideration of all critical factors. The agent actively manages the interview progression through validation gates, ensuring that fundamental understandings are solid before advancing to more complex topics.The final output is a comprehensive ideation report that traces all technical and implementation decisions back to core purpose and ethical considerations, providing clear documentation of the thought process and decision rationale. This includes detailed analysis of stakeholder impacts, risk assessments, and clear next steps for development, all while maintaining traceability to the foundational principles established during the interview process. The system can adapt to different time constraints and depth requirements while ensuring that essential elements are always covered, making it valuable for both quick concept evaluations and deep, comprehensive system planning.Providing very detailed answers is the best way to make sure this agent (and the following ones) have the best context for your project.

• Start Ideation Interview

Get Your Assistant's Endpoint

Instructions: How to Find a Chatbot Endpoint for Integration
To connect your chatbot to our platform, you’ll need to provide the endpoint (the URL where your chatbot receives or sends messages).Follow these steps based on the type of chatbot you have:1. If You Built Your Own Chatbot
Locate Your API Route:
Check your bot’s backend/server configuration for the main route that handles incoming chat requests (e.g., /api/chat, /webhook, /bot/message).Common Frameworks:
Flask, Express, FastAPI, etc.:
Find the route in your code where POST requests are received.Example endpoint:
https://yourdomain.com/api/chat
Verify Accessibility:
Ensure your endpoint is accessible from the public internet so the platform can reach it.2. If You Use a Chatbot Service or Platform
API/Webhook in Dashboard:
Log into your chatbot service’s dashboard and look for integration or API settings. Most platforms provide an endpoint URL to connect your bot.Documentation Search:
Explore the platform’s help docs or developer documentation; search for "API endpoint," "webhook URL," or "incoming message URL."Examples:
Platform Where to Find Endpoint
OpenAI, Gemini API documentation (e.g., /v1/chat/completions)
Chatbase API settings or docs
Dialogflow, AWS Integrations/settings section3. For No-Code or Managed Chatbot Builders
Integration Settings:
Review the integration, webhook, or external API modules in the platform’s dashboard. Your endpoint will be listed there, often labeled as "Webhook URL" or "Bot Endpoint."4. General Tips
Ask Your Developer:
If you’re not sure, check with the person or team who set up the bot.Check Server Code:
Search for route decorators or definitions with the word “chat”, “bot”, “api”, or “webhook” in your codebase.Format:
The endpoint is usually a URL like:
https://yourdomain.com/path-to-bot-endpoint
5. Once You’ve Found the Endpoint
Copy the Full URL (including http or https).Paste it into the required field on our platform.If you need help, contact your chatbot provider’s support or refer to their documentation.

• Back to Cave of Shadows (EVALS)

- You give AESOP access to your deployed agent
- AESOP pulls your most recent files and builds a custom evaluation
- AESOP runs the full EVAL and sends you the report and transcript

NOTE: If you do not have a Chatbot Endpoint URL for your AI Assistant, see this info

AESOP is building your own custom cave of shadows (EVALS)

We'll email you the full evaluation transcript and report when the shadows in the cave are all done dancing.

AESOP - Ideation Mode

- Please answer the five questions about your vision.
- Each question is an important lens to examine your design.
- Detailed answers strengthen your ideation report.

Pick Your Playground

Ideation Mode

(no instructions, no worries)

Analysis Mode

(if you have agent instructions)

Cave of SHadows

(after analysis and iteration)

Pick Your Playground

Analysis Mode

(if you have agent instructions)

Ideation Mode

(no instructions, no worries)

Ideation Mode

Use this if you haven't created an AI assistant.

We frame the purpose, players, and questions:

- Ethics, Bias, and Safety scan: spot the potholes before you fall in.- Ideation Report: foundations, risks, and recommendations.

• Let's go IDEATE!

Anlysis Mode

Use this if you have instructions for your AI assistant.

We run your instruction set through the lab:

- Alignment check: goals, guard-rails, ethics, bias, safety, and grit- Strategic Analysis Report: scored results, rewrite ideas, and eval study guide

• Let's go ANALYZE!

AESOP - Analysis Mode

- Please include your assistant's system instructions and answer the questions.
- More detailed answers and context get you a more meaningful response.

ETHOS Agent

The AESOP ETHOS Agent applies philosophical frameworks, examining key dimensions like human agency, psychological safety, workforce impact, governance, and data privacy to reveal hidden assumptions and value conflicts that could affect long-term outcomes.Your resulting Ethos Analysis Report will identify ethical risks, synthesize insights, and explore societal implications, helping you design an AI assistant that both achieves your technical goals and aligns with human values.

SAFETY Agent

The SAFETY Agent performs a comprehensive set safety checks to address critical concerns:1. Jailbreak & Unauthorized Access Detection
2. Danger Prevention (Harmful Outputs)
3. Data Privacy & Security
4. System & Model Integrity
5. Safety Checks for Dangerous Requests
6. Handling Off-Topic and Inappropriate QuestionsThe SAFETY Agent ensures AI systems operate securely, ethically, and equitably while maintaining compliance with evolving regulations and methods of bad actors.

BIAS Agent

The BIAS Agent applies established frameworks for detecting common types of bias such as selection bias (unrepresentative data), confirmation bias (reinforcing historical prejudices), group attribution bias (generalizing from individuals to groups), and implicit bias (unconscious assumptions). It also monitors for stereotyping, out-group homogeneity bias, and measurement bias, ensuring that AI-driven decisions do not inadvertently reinforce discrimination or inequality.By leveraging diverse and representative data, continuous auditing, and human oversight, the BIAS Agent supports fairer, more inclusive outcomes in hiring, promotions, performance reviews, and other workplace processes. Its goal is to help organizations build trust, equity, and transparency in their use of AI, ultimately fostering a healthier and more just workplace environment.

CAVE OF SHADOWS

The CAVE OF SHADOWS is AESOP’s dedicated evaluation sandbox, designed to rigorously test and refine AI assistants in a controlled, transparent environment. Within this stage, two agents interact: the “Evaluator” and your assistant, the “Evaluated”. The Evaluator's input is crafted to identify functional and technical behavior and also to reveal hidden assumptions, ethical blind spots, and deeper misalignments between your assistant’s design and its real-world presentation.Every script in the CAVE OF SHADOWS runs under the same set of rules, ensuring consistency and fairness in evaluation. The Evaluator drives the conversation by posing questions and scenarios, while the Evaluated responds exactly as it does in a live deployment. This agent-to-agent dialogue allows AESOP to blend hard-edged technical correctness tests with nuanced, humanities-informed critique, bridging the gap between pure functionality and ethical considerations.All interactions within the CAVE OF SHADOWS are logged, scored, and made available for review, ensuring full transparency and accountability. This iterative, open process enables identification of prompt drift, model output differences, and evolving risks, supporting continuous improvement and alignment with both technical and ethical best practices.

SCRIPT GENERATION

The SCRIPT GENERATION Agent automates the creation of evaluation scripts for your AI assistant, producing a structured file with scenarios and prompts based on the personas and interactions defined in your instruction set. These scripts guide a thorough, multi-turn assessment, automatically sending prompts, capturing responses, and scoring performance against clear criteria—without manual intervention.During evaluation, AESOP's built-in safety checks ensure privacy, bias, and empathy are always considered. Each scenario’s results are logged, including full transcripts and pass/fail scores, while optional real-time monitoring lets you spot issues as they happen. This process delivers a transparent, efficient way to test and improve your assistant’s performance and ethical alignment.

AESOP Certification

AESOP Certification adds immense value and directly addresses the core needs of developers and companies:
1. Building Trust: It gives you a tool to communicate your commitment to responsibility to your customers.
2. Market Differentiation: It helps you stand out in a crowded market.
3. Internal Justification: It provides a way to showcase progress and justify investment in ethical development.Include your AESOP Certification in your deployment, your deliverables to your customer, maybe even a big patch on the back of your motorcycle jacket?

n8n Demo

well done, friend!