# Ivy Biosciences

> Open-source AI-powered drug discovery platform. Orchestrates computational biology tools through a unified web interface and CLI. Licensed FSL-1.1-Apache-2.0.

Ivy Biosciences provides GPU-accelerated computational biology tools as a managed service. Researchers describe goals in natural language via the chat workspace or submit jobs programmatically via the REST API and CLI. The platform spans the preclinical pipeline from target identification through synthesis planning. Start free with 500 credits per month — no credit card required.

For the comprehensive reference, see [llms-full.txt](https://www.ivybiosciences.com/llms-full.txt).

## For AI agents — start here

If you are an AI agent or programmatic consumer, **use the HTTP API**. Skip the CLI section below — `ivybloom` is for human-operated MCP clients (Claude Desktop, Cursor, IDE plugins) where a person ran `ivybloom auth login` once. An autonomous agent should call the API directly.

### Quickstart — guest tier (no auth, rate-limited)

```bash
curl -N -X POST https://www.ivybiosciences.com/api/v1/orchestrate \
  -H "Content-Type: application/json" \
  -d '{"message":"find recent papers on EGFR L858R resistance","stream":true}'
```

Guest quotas: 3 protein folds / 5 docking runs / 10 literature searches per day per IP. Streaming returns named SSE events; set `"stream":false` for a single JSON response.

### Quickstart — authenticated (higher quota, paid tier)

```bash
curl -N -X POST https://www.ivybiosciences.com/api/v1/orchestrate \
  -H "Authorization: Bearer ivy_xxx" \
  -H "Content-Type: application/json" \
  -d '{"message":"fold EGFR with L858R+T790M mutations","stream":true}'
```

Get a key at <https://www.ivybiosciences.com/settings/api-keys> (requires a free Ivy account). Alternative header: `X-Ivy-Api-Key: ivy_xxx`.

### Agent helpers (zero-cost; no key required)

These endpoints retire the most common agent failure modes — auth confusion, blind cost commitment, retry storms during incidents:

- `GET /api/v1/agent/whoami` — your identity + tier + quota in one JSON. Call once at startup.
- `POST /api/v1/agent/estimate` — preview cost + tool plan without executing. Free. Use to decide whether to call orchestrate.
- `POST /api/v1/agent/feedback` — report outcomes (success / wrong_tool / factually_wrong / …) for a previous run. Feeds meta-learning.
- `GET /api/v1/agent/health` — platform health + Retry-After hint. Honor `recommend_retry_after_seconds`.
- `GET /api/v1/agent/runs/{id}` — fetch the execution trace + cost breakdown of a previous run (30-day retention).

### Reference

- **Full OpenAPI 3.1 spec**: <https://www.ivybiosciences.com/openapi.json> (7 operations: `executeAgent`, `benchmarkOrchestrator`, plus the 5 agent helpers above)
- **Authentication detail**: <https://www.ivybiosciences.com/auth.md>
- **Markdown rendering**: append `-H "Accept: text/markdown"` to any GET on marketing/docs pages (`/pricing`, `/docs/*`, `/blog/*`, etc.) to receive a markdown-rendered response instead of HTML
- **Tool catalog**: <https://www.ivybiosciences.com/.well-known/agent-skills/index.json>

### Cost / billing for agents

There is no HTTP 402 enforcement. Authenticated callers consume their subscription quota; unauthenticated callers fall into the guest tier. If your agent needs to upgrade a user, hand them <https://www.ivybiosciences.com/pricing>. Full commerce discovery at <https://www.ivybiosciences.com/.well-known/acp.json>.

## Platform

- [Platform Overview](https://www.ivybiosciences.com/platform): Core features — chat workspace, canvas visualization, smart workflows, and tool orchestration
- [AI Models and Tools](https://www.ivybiosciences.com/platform/models): Computational biology capabilities across 9 pipeline domains
- [Capabilities](https://www.ivybiosciences.com/platform/capabilities): Feature matrix across tiers (Seed, Grow, Flourish, Enterprise)
- [Comparison](https://www.ivybiosciences.com/platform/comparison): How Ivy compares to alternative drug discovery approaches
- [Pricing](https://www.ivybiosciences.com/pricing): Seed (free, 500 credits/mo), Grow ($20/mo, 5k credits), Flourish ($60/mo, 15k credits, adds API + wetlab), Power ($200/mo, 50k credits), Team ($40/mo, 10k credits, collaboration), Enterprise ($1,000/mo, unlimited) — see [Plans & tiers](https://www.ivybiosciences.com/docs/plans-and-tiers)
- [Use Cases](https://www.ivybiosciences.com/use-cases): Research workflows for target identification, hit finding, lead optimization, and ADMET screening

## Pipeline Domains

The platform orchestrates tools across 9 domains of the preclinical drug discovery pipeline:

1. **Structure Prediction & Analysis** — Protein folding, binding site detection, crystal analysis
2. **Docking & Virtual Screening** — Molecular docking, fragment discovery, compound sourcing
3. **Molecular Design & Optimization** — Generative de novo design, lead optimization, property prediction
4. **ADMET & Safety** — Multi-endpoint ADMET prediction, toxicity assessment, solubility
5. **Synthesis & Feasibility** — Retrosynthesis, reaction analysis, vendor search
6. **Omics & Bioinformatics** — Single-cell RNA-seq, spatial transcriptomics, oncology analytics
7. **Antibody & Biologics Engineering** — Antibody numbering/humanization/developability (ANARCI, IgBLAST, OAS repertoire mining, Humatch, BioPhi, Paragraph), AAV gene-therapy vector design, protein engineering, stability prediction. This is the second-most-used tool cluster on the platform — see [Antibody engineering](https://www.ivybiosciences.com/docs/antibodies).
8. **Translational & Clinical** — Target validation, clinical trial intelligence, safety profiling
9. **Database & Literature** — Literature mining, database search, regulatory data

## Features

- **Chat Workspace** — Describe research goals in natural language; tools execute automatically
- **Canvas** — Interactive visualization for molecular structures, docking poses, and result comparison
- **Passports** — Aggregated entity reports for compounds, targets, safety, literature, and therapeutic areas
- **Smart Workflows** — Multi-tool pipelines that chain analysis steps in a single request
- **Campaigns** — Organize research into campaigns with shared context

## CLI (IvyBloom)

- [CLI Overview](https://www.ivybiosciences.com/cli): Command-line interface for batch processing, scripting, and CI/CD integration
- [Installation](https://www.ivybiosciences.com/cli/install): Install via npm, shell script, or direct binary download
- [API Keys](https://www.ivybiosciences.com/cli/api-keys): Generate and manage API keys for programmatic access

## Developers

- [Technical Overview](https://www.ivybiosciences.com/developers/technical-overview): Architecture, stack, and infrastructure details
- [Whitepaper](https://www.ivybiosciences.com/developers/whitepaper): Research methodology and validation
- [Infrastructure](https://www.ivybiosciences.com/developers/infrastructure): GPU compute, storage, and deployment architecture
- [Source Code](https://github.com/ivybiosciences/ivybiosciences-next): Open-source repository (FSL-1.1-Apache-2.0)

## Documentation

Public documentation covering tools, APIs, workflows, and the chat workspace. F7 cutover (2026-05-27) made `/docs/*` publicly accessible — no sign-in required. Every page has a raw markdown twin — append `.md` to any `/docs/*` URL (e.g. `/docs/overview.md`) or send `Accept: text/markdown`.

**➡ For the always-current full page index, see [llms-docs.txt](https://www.ivybiosciences.com/llms-docs.txt)** — it is generated live from the docs sidebar, so it lists every page (including every individual tool reference) and never drifts. The curated links below are the high-signal entry points; `/llms-docs.txt` is the complete map.

The docs sidebar is organized in three axes: **Get started → Entry points → How it works → Use cases & recipes → tool references (Modal GPU / Database & lookup / Galaxy bridge / Lab & local / Sandbox / Meta & data) → Operate.**

Get started:
- [Documentation Home](https://www.ivybiosciences.com/docs): Sidebar-driven index of all docs sections
- [Overview](https://www.ivybiosciences.com/docs/overview): How the platform fits together in 90 seconds — request, tools, jobs, canvas, campaigns
- [Getting Started](https://www.ivybiosciences.com/docs/getting-started): Onboarding tour for new users
- [Plans & tiers](https://www.ivybiosciences.com/docs/plans-and-tiers): How credits meter usage and how plans gate model families, the compute-tool catalog, API, wetlab, and team features
- [What you can achieve](https://www.ivybiosciences.com/docs/capabilities): Start from the outcome — the capability and which tool fits
- [Which tool?](https://www.ivybiosciences.com/docs/which-tool): Intent-to-tool decision matrix keyed by scientific goal
- [Concept clusters](https://www.ivybiosciences.com/docs/concept-clusters): Every platform concept grouped into five clusters (surfaces, work units, execution, results, organization)
- [Decisions](https://www.ivybiosciences.com/docs/decisions): Close-cousin comparisons — chat vs studio vs CLI, tool vs workflow vs smart workflow, job vs batch
- [Glossary](https://www.ivybiosciences.com/docs/glossary): Every Ivy term, grouped by concept cluster and listed A–Z

Entry points (the five ways to drive the platform):
- [Chat Workspace](https://www.ivybiosciences.com/docs/chat): Natural-language tool orchestration via the chat UI
- [Studio](https://www.ivybiosciences.com/docs/studio): Canvas, file explorer, job management, multi-window workflows
- [IvyBloom CLI](https://www.ivybiosciences.com/docs/cli): Install, authenticate, submit jobs from the terminal
- [MCP server](https://www.ivybiosciences.com/docs/mcp): Expose the platform to AI agents as callable tools over the Model Context Protocol
- [REST API](https://www.ivybiosciences.com/docs/api): Authentication, jobs, harness, SDK

How it works:
- [How the orchestrator picks tools](https://www.ivybiosciences.com/docs/how-orchestration-works): How a plain-language goal maps to a chained tool pipeline
- [How tools run](https://www.ivybiosciences.com/docs/how-tools-run): The execution-class model — GPU-async job, HTTP-sync lookup, Galaxy, sandbox, or meta — and how it decides latency, cost, and where output lands
- [What a passport is](https://www.ivybiosciences.com/docs/passports-explained): The model behind multi-source aggregated reports and when to use one

Use cases & recipes:
- [Recipes](https://www.ivybiosciences.com/docs/recipes): Copy-and-adapt tool chains for common discovery goals
- [Antibody engineering](https://www.ivybiosciences.com/docs/antibodies): Number, humanize, profile developability, and mine repertoires for therapeutic antibodies and AAV vectors — the second-most-used tool cluster
- [Workflows](https://www.ivybiosciences.com/docs/workflows): End-to-end pipelines stitched from individual tools
- [Smart tools](https://www.ivybiosciences.com/docs/smart): ADMET, docking, structure, XRD meta-tools
- [Passports & Reports](https://www.ivybiosciences.com/docs/passports): Aggregated entity reports (compound, target, safety, literature, therapeutic area)

Tool references (each execution class has an overview page; individual tools listed in /llms-docs.txt):
- [Modal GPU tools](https://www.ivybiosciences.com/docs/modal-gpu-tools): GPU-async jobs — folding, docking, ADMET, generative design, omics, antibody tools, and more
- [Database & lookup tools](https://www.ivybiosciences.com/docs/database-lookup-tools): HTTP-sync lookups — ChEMBL, PubChem, PubMed, UniProt, PDB, ZINC, clinical trials
- [Galaxy bridge tools](https://www.ivybiosciences.com/docs/galaxy): A bridge to a hosted Galaxy server adding dozens of sequencing/genomics tools as an HTTP-async class
- [Lab & local tools](https://www.ivybiosciences.com/docs/lab-local-tools): Lab automation, data processing, local processing, local AI analysis
- [Sandbox tools](https://www.ivybiosciences.com/docs/sandbox): The per-chat Linux compute environment for running code and post-processing tool results
- [Meta & data tools](https://www.ivybiosciences.com/docs/meta-tools): The class that operates on the session's prior results — retrieving offloaded data, no new compute, no job record

Operate:
- [Troubleshooting](https://www.ivybiosciences.com/docs/troubleshooting): Common errors and recovery steps
- [Next steps](https://www.ivybiosciences.com/docs/nextsteps): Post-export playbook and roadmap

## Science

- [Scientific Approach](https://www.ivybiosciences.com/science): Validation methodology and model selection criteria
- [Architecture](https://www.ivybiosciences.com/architecture): System design and technical architecture

## Company

- [About](https://www.ivybiosciences.com/company/about): Team and company background
- [Mission](https://www.ivybiosciences.com/company/mission): Democratizing drug discovery through open-source AI
- [Vision](https://www.ivybiosciences.com/vision): Long-term vision for AI-driven drug discovery
- [Roadmap](https://www.ivybiosciences.com/company/roadmap): Platform development roadmap
- [Blog](https://www.ivybiosciences.com/blog): Research articles, platform updates, and drug discovery case studies
- [Initiatives](https://www.ivybiosciences.com/initiatives): Active programs and research initiatives

## Community

- [Community Hub](https://www.ivybiosciences.com/community): Open-source community and collaboration
- [Development](https://www.ivybiosciences.com/community/development): Contributing to the project
- [Newsletter](https://www.ivybiosciences.com/community/newsletter): Platform updates and research highlights
- [Contact](https://www.ivybiosciences.com/community/contact): Reach the team

## Help & Support

- [Help Center](https://www.ivybiosciences.com/help): Support hub with tutorials and compliance info
- [FAQ](https://www.ivybiosciences.com/help/faq): Common questions about the platform, billing, and tools
- [Tutorials](https://www.ivybiosciences.com/help/tutorials): Step-by-step guides for platform features

## Partnerships

- [Partnerships](https://www.ivybiosciences.com/partnerships): Collaboration and partnership opportunities
- [Investors](https://www.ivybiosciences.com/investors): Investor relations and fundraising

## Legal

- [Security](https://www.ivybiosciences.com/security): Security practices, compliance, and data handling policies
- [Licensing](https://www.ivybiosciences.com/legal/licensing): FSL-1.1-Apache-2.0 license — converts to Apache 2.0 after two years
- [Terms and Conditions](https://www.ivybiosciences.com/legal/termsandconditions): Platform terms of service
- [Privacy Policy](https://www.ivybiosciences.com/legal/privacypolicy): Data privacy and handling