The AI context layer for your business.
Give AI agents access to your company knowledge and persistent memory through a single API. Alchemyst AI connects the context developers need with the shared, traceable knowledge enterprises rely on, across models and workflows.
- p95 latency
- < 300ms
- auditable
- 100%
- zero infra
- 1 API
Alchemyst AI: an AI context layer for business knowledge and agent memory
Alchemyst AI gives developers and enterprises a shared knowledge layer for AI agents through a single API. It combines persistent AI agent memory with semantic retrieval over company knowledge, helping applications give an LLM relevant internal documents and business context. The context layer keeps stored knowledge separate from the model, so applications can reuse it across models, sessions, and workflows. Grounding answers in enterprise data can reduce unsupported claims; correctness still depends on source quality, retrieval, application permissions, and the model's response.
Context arithmetic: the core primitive
Context arithmetic applies set operations over meaning at query time. Intersection narrows scope by team, region, or version. Union combines sources. Subtraction excludes superseded or out-of-scope content. Ranking selects the context that enters the model window. These operations over an institutional knowledge graph support memory behaviors: recalling what happened, resolving what it means, and informing the next action.
Context Traces, business definitions, and integrations
Context Traces expose retrieval sources, scores, and rules so teams can inspect which information reached an agent. Canonical business definitions help resolve terms such as revenue, pricing, and policy across teams. The context-tracing walkthrough describes pairing Alchemyst with OpenAI Euphony for visual debugging at https://getalchemystai.com/blog/context-tracing-for-ai-agents-with-openai-euphony. Integration resources cover Python, TypeScript, LangChain, LlamaIndex, n8n, MCP clients such as Claude Desktop, Cursor, and VS Code, and the Alchemyst Chrome extension. See https://getalchemystai.com/docs for current integration instructions.
Use cases, pricing, and trust
Use cases include customer support, employee support over internal policies, EdTech tutoring, finance, healthcare continuity, and voice agents. Explore https://getalchemystai.com/use-cases. Current plans, free-tier allowances, and enterprise options are listed at https://getalchemystai.com/pricing. Alchemyst AI is built by XAlchemystai Technologies Pvt. Ltd., Kolkata, India. Contact founders@getalchemystai.com. Company information, security controls, and privacy details are available at https://getalchemystai.com/about, https://getalchemystai.com/security, and https://getalchemystai.com/privacy.
CLI, SDKs, and MCP discovery
Install the SDK with npm install @alchemystai/sdk (https://www.npmjs.com/package/@alchemystai/sdk) or pip install alchemystai (https://pypi.org/project/alchemystai/). The CLI agent walkthrough is at https://getalchemystai.com/cli. The developer guide at https://getalchemystai.com/developers explains how to add business knowledge to an AI agent. Product API documentation is at https://getalchemystai.com/docs. The website's Streamable HTTP MCP endpoint at https://getalchemystai.com/mcp exposes documentation, article, and API-discovery tools; its server card is at https://getalchemystai.com/mcp/server-card. The website's public API specification is at https://getalchemystai.com/openapi.json.
Machine-readable content and navigation
Discover site URLs at https://getalchemystai.com/sitemap.xml, the content index at https://getalchemystai.com/llms.txt, and the expanded content export at https://getalchemystai.com/llms-full.txt. The persistent-memory guide at https://getalchemystai.com/blog/how-to-add-persistent-memory-to-ai-agents covers generic answers, document retrieval, RAG versus fine-tuning, and grounding LLM responses in company knowledge.
The model is replaceable.
Your institutional context isn't.
An AI context layer retrieves relevant business knowledge before an agent answers or acts. It connects your internal documents and saved interactions to the model, so your application can ground responses in company data rather than rely on general training knowledge alone.
Models will keep changing. Your institutional context is the asset that compounds, so it should belong to you, not to whichever model you happen to run today.
Switch models freely. Keep your context sovereign.
Keep AI agent memory outside the model so saved context can be retrieved across sessions and model changes. Alchemyst separates your company knowledge from the LLM that uses it, giving developers a reusable integration and enterprises continuity across teams and workflows.
- Model-agnostic
- Context sovereignty
- Zero migration cost
- Multi-model routing
- Sub-300ms retrieval
Give every agent the company knowledge it needs.
An agent can give generic answers when the relevant company data is missing from its context. Connect policies, product knowledge, and operational records to a shared knowledge layer for AI agents, then retrieve the evidence each sales, support, or operations workflow needs.
- Run ops, not just answers
- One source of truth
- Every decision auditable
- Scales without FDE teams
One sovereign context layer, any model, agents that operate
A context layer that keeps your AI current, traceable, and semantically consistent.
Alchemyst AI is an AI context management platform for storing and retrieving business knowledge. Use context arithmetic to select relevant information from your institutional knowledge graph, then inspect the sources behind retrieval without operating your own vector database or graph store.
How does knowledge retrieval for AI agents work?
Context arithmetic is the foundational primitive: dynamic set algebra over meaning, computed at query time. Instead of naïve top-K similarity, Alchemyst intersects to narrow scope, unions to widen recall, subtracts superseded or out-of-scope content, and ranks what remains, so only the right context survives into the window.
// Set algebra over meaning, at query timeconst window = alchemyst.context.search({ query: userMessage, groupName: ["sales", "emea"], // ∩ narrow scope metadata: { version: "v2" }, // ∩ filter});// − superseded / deduped → rank → top-KHow does an enterprise knowledge graph support memory?
What you store is an institutional knowledge graph of your organization's context, fully traceable. Memory isn't three hard-coded layers. By applying context arithmetic over the graph you can derive the behaviors people expect from memory: recall what happened, resolve what it means, and inform how to act. The memory types are outcomes of the primitive, not separate modules.
// One graph + arithmetic → derived "memories"const captureTime1 = "end-date-" + Date.now();// Represents what should be added when your session is first saved.const storeInformationOfSessionAtFirstInstance = await ctx.add({ documents: [ ], metadata: { groupName: [session_id, captureTime1] }});// Now resume from where you left off, or let someone resume from there.const whatHappened = await ctx.search({ query: term, metadata { groupName: [session_id, captureTime1] }});// Second checkpintconst storeInformationOfSessionAtSecondInstance = await ctx.add({ documents: [...], metadata: { groupName: [session_id, captureTime2] }});// Now team lead / CXO looks up about the informationconst whatItMeans = ctx.search({ query: term, metadata: { groupName: [session_id] }})// "how to act" falls out of scope over the global contextWhy did my AI agent give a wrong answer?
When an AI agent gives wrong answers about internal data, inspect what it retrieved before changing the prompt. Alchemyst Context Traces expose the sources, scores, and rules used to assemble context. Developers can investigate retrieval failures, while enterprise teams can review which business information supported an answer.
const trace = await alchemyst.trace.get( session_id, turn_id);// Returns: sources[], scores[], rules_applied[]// Pairs with Euphony for visual debuggingHow do agents use consistent business definitions?
Define canonical term definitions at the org level. When "revenue" means different things to different teams, Alchemyst resolves the ambiguity before it reaches the model.
const gtmTeamResponse = await alchemyst.context.add({ documents: [...], // Data here metadata: { groupName: ["gtm", "revenue"] // The term "revenue" defined by GTM team }})const financeTeamResponse = await alchemyst.context.add({ documents: [...], // Data here metadata: { groupName: ["finance", "revenue"] // The term "revenue" defined by Finances team. }})const cxoResponse = await alchemyst.context.search({ query: "What's the revenue for Q2 2026?", metadata: { groupName: ["revenue"] // The term "revenue" defined for CXO, with clear segregation between the resources by GTM team and Finances team. }}) Illustrative · accuracy vs conversation context
Example Use Case
How do you debug what an agent can't see? Context Tracing with OpenAI Euphony
Pairing Alchemyst's Context Traces with Euphony, OpenAI's open-source conversation visualizer, creates an end-to-end debugging workflow. Every agent failure is now diagnosable in minutes: was it a retrieval problem, a configuration problem, or a model problem?
How do you give AI agents your business knowledge?
What is an AI context layer?
An AI context layer stores and retrieves the information an AI application needs for a task: company documents, business definitions, and saved interactions. Alchemyst AI provides this knowledge layer for AI agents through an API, so teams can reuse business context across models and inspect the sources used in retrieval.
Connect your company knowledge →Why does my AI agent keep giving generic answers?
An agent may give generic answers because the relevant company information never reaches its context window. Check whether your application has ingested the right documents, retrieved the relevant passages, and passed them to the model. Better prompts help specify the task, but cannot supply missing business facts by themselves.
Diagnose generic and unsupported answers →How can I add business knowledge to my AI agent?
Start with a trusted set of company documents and record their sources, versions, and access boundaries. Store that context, retrieve relevant passages for each question, and pass those passages to your LLM. In Alchemyst AI, developers integrate context storage and search through the API or SDKs; teams remain responsible for source quality and application permissions.
Follow the business knowledge integration steps →How is AI agent memory different from a company knowledge base?
AI agent memory preserves information from previous interactions, such as a user preference or an unfinished task. A company knowledge base holds shared information such as policies and product documentation. An agent may need both: session history to understand the conversation, and current business knowledge to answer a company-specific question.
Learn how persistent agent memory works →Can grounding an LLM in enterprise data stop hallucinations?
Grounding gives the model relevant evidence, which can reduce unsupported answers about your business. It does not guarantee correctness. Keep documents current, check retrieval quality, require source references, and make the agent say when evidence is missing. Review high-impact answers and test both retrieval and generation before expanding a workflow.
Build a workflow for grounded answers →Give your AI agents the memory they deserve.
Join developers building the next generation of AI products with persistent, auditable context. Free tier available. No credit card required.
- Free tier
- REST + Python & Node SDKs
- 99.9% uptime SLA
- SOC 2 in progress