Last updated: April 5, 2026
SuperAgentic is an agentic financial research system. Unlike a generic chatbot, the agent comes equipped with a curated library of professional-grade knowledge blocks: tools (live data connectors), skills (structured analytical workflows like DCF valuation or Altman Z-Score), and sources (vetted data providers).
The agent is aware of all its capabilities at all times — you do not need to learn them, remember them, or invoke them by name. Just describe your research question in plain language and the agent assembles the right workflow on the fly.
Everything is transparent. Every tool call, every data source, every skill invocation is visible in the conversation — click any tool card to inspect the raw data, or open the sources sidebar to see exactly which providers contributed to an answer. Nothing happens behind a black box.
Most importantly, SuperAgentic adapts to you. Write your own custom skills to encode your personal investment methodology — your own analytical frameworks, the way you think about risk, the screens you run. Upload your own files as personal data sources that the agent can read, analyze, and cross-reference with live market data. Set up a custom briefing that teaches the agent your investment style, focus areas, and preferences. Save frequently used prompts as custom commands for faster access. The platform becomes more powerful the more you make it yours.
When you ask a question, the agent decides which tools and skills to invoke, fetches live data from multiple providers, runs analytical models (including Python computations in a secure sandbox when needed), and synthesizes everything into a single, cited answer. Tool result cards appear above the text answer so you can inspect the raw data yourself.
You can click any tool card to see exactly what data was fetched. The sources button (globe icon) on each message opens a sidebar showing every data provider and web link used in that answer.
SuperAgentic connects to 15 vetted data providers covering a wide range of financial and economic data. All data is fetched live — no stale caches. Here is the scope of what the agent can access:
Beyond raw data, SuperAgentic knows how to think about it. The agent has access to over 60 structured analytical workflows drawn from professional finance — the same methodologies used by equity analysts, portfolio managers, and hedge funds.
Ask whether a stock is manipulating earnings and the agent runs a Beneish M-Score. Wonder if your retirement portfolio can survive a bear market? It launches a Monte Carlo simulation with 5,000 paths. Want to screen for undervalued quality stocks the way Joel Greenblatt would? There's a Magic Formula skill for that. Need to stress-test a portfolio allocation against historical crises? The agent knows how.
Many of these skills run real Python computations in a secure sandbox — producing auditable numbers, not LLM guesses. You don't need to know the methodology names. Just describe what you want to understand, and the agent picks the right approach.
Explore the full catalog on the Resources page.
Advanced skills run Python computations in a secure, sandboxed environment. Monte Carlo simulations, scoring models, and statistical tests produce auditable, precise results — not LLM arithmetic.
Upload spreadsheets, CSVs, PDFs, and documents to your personal “My Data” workspace. The agent can read, analyze, and modify your files — generating new output files with full version history. Original files are never overwritten.
Generate professional equity research reports from any conversation. Click the report icon on any message to export a branded, downloadable PDF. View all your reports in the “My Reports” tab.
Financial charts render directly in the chat. Charts are exportable as images.
Star important conversations for quick access. Create custom color-coded tags (up to 20) to organize conversations by topic, project, or priority. Filter conversations by stars or tags from the sidebar.
The empty chat screen shows curated suggestion cards to help you discover capabilities. Dismiss cards you don't need — they won't come back unless you restore them from Settings.
Type / in the chat input to open a searchable palette of all tools and saved prompts.
No need to compromise. SuperAgentic is provider-agnostic — our system runs over any frontier LLM, so you can choose the model you trust most. Use the provider selector in the top bar to switch at any time.
Every model gets access to the same tools, skills, and data sources. We continuously integrate the latest advancements in generative AI, so the platform keeps getting smarter without you having to change anything.
SuperAgentic uses a credit-based system. Each credit is worth $0.01 USD. New accounts receive a starting balance. Credits are deducted based on the AI processing required for each request (tokens used, data API calls, code execution).
Your credit balance is shown in the top bar. When your balance reaches zero, new requests are paused until credits are added.
Some analyses require multiple processing steps that the agent chains automatically. If a task needs more processing than your remaining credits allow, we finish the job before reconciling your balance — so it may go briefly negative. This guarantees you always receive the complete results of work you were charged for. No credits are ever spent without delivering a work product.
Data is fetched live from providers when you ask — there is no stale cache. Stock quotes may have a 15-minute delay (free tier standard). Financial statements, macro data, and crypto prices are as current as each provider makes them.
Yes. Conversations are stored securely and are only accessible to your account. Uploaded files are stored in a private cloud bucket tied to your user ID. You can export or delete your data at any time from Settings. See our Privacy Policy for full details.
Supported formats: Excel (.xlsx, .xls), CSV, PDF, Word (.docx), plain text, Markdown, and JSON. Maximum 10 MB per file, 20 files and 50 MB total per account.
Yes. The agent can read your files and create modified versions using Python (pandas, openpyxl). Your original file is never overwritten — the agent always creates a new derived file. Both appear in My Data with full version history.
For complex calculations (DCF models, Monte Carlo simulations, statistical tests), the agent writes and runs Python code in a secure sandbox. This produces auditable, precise results rather than relying on LLM arithmetic. The sandbox has no network access and runs with pre-installed libraries (numpy, pandas, scipy, scikit-learn, statsmodels, openpyxl).
Yes. Go to Resources and click “Create Skill.” You can write a prompt template from scratch, fork an existing system skill to customize it, or save a quick prompt shortcut. You can also submit your skills for admin review to share them with the community.
Complex analyses sometimes span multiple processing steps that the agent chains automatically. If the task requires more processing than your remaining balance allows, we let the agent finish rather than cutting it off mid-work. Your balance may go briefly negative as a result. This ensures every credit you spend produces a complete deliverable — no partial results, no wasted charges. Your next credit purchase restores a positive balance.
Browse the full catalog of tools, skills, and data sources on the Resources page. For questions not covered here, contact us at support@superagentic.ai.