Tara

Query Tahoe’s frontier-scale datasets from your AI agent.

Available today: Tahoe-100M, a 100-million-cell perturbation atlas. Ask in plain language for expression and differential expression, pathway scores, phenotype readouts, embedding distances, perturbation markers and drug and cell-line annotations, or run guarded read-only SQL — no downloads, just one install and a GitHub sign‑in.

Watch the demo

Paste it into your AI agent to install Tara.

Two examples. In Codex, Tara is installed with the onboarding prompt, then compares adagrasib and RMC-6236 responses across six KRAS-G12C models in Tahoe-100M and builds an interactive dashboard of binding pose and transcriptional response. Then, in Claude Code, it finds the compounds whose pathway responses most resemble saquinavir's and writes the findings up as a webpage. In Cursor, it answers whether binimetinib down-regulates KRAS signaling and writes the answer to a file.
Same plugin in
Install

One install, then sign in with GitHub.

You'll be asked to sign in with GitHub. If tools don't appear, restart your session.

Ask your agent

Paste this into your agent:

Install the Tahoe plugin: add the marketplace tahoebio/tahoe-public-plugins, install the tara plugin, then sign me in with GitHub.

Claude Code

In Claude Code:

claude plugin marketplace add tahoebio/tahoe-public-plugins
claude plugin install tara@tahoe-public

Codex

In Codex:

codex plugin marketplace add tahoebio/tahoe-public-plugins
codex plugin add tara@tahoe-public

Advanced

Connect the server directly, without the plugin:

claude mcp add --transport http --scope user tahoe-public https://tara.tahoebio.io/mcp
codex mcp add tahoe-public --url https://tara.tahoebio.io/mcp
Examples

Ask the open-ended question. Get the analysis back.

Tara works alongside the other tools in your agent, so a question about Tahoe’s data can end in a dashboard or a write-up, not just a number.

Tools

What you get

  • list_releases

    The dataset release this service serves. Call this first.

  • search_conditions

    Find conditions — drug, dose and cell line — in the index.

  • resolve_genes

    Resolve gene symbols and Ensembl ids; exact, ambiguous or not_found, never a guess.

  • list_cell_lines

    The cell lines: Cellosaurus id, name, DepMap id, organ and driver annotations.

  • get_drug_info

    Drug annotations: targets, mechanism class, approval status; filter by drug, moa or target.

  • get_expression_slice

    Pseudobulk expression for genes × conditions, with matched controls and the release and request it answered.

  • get_differential_expression

    Drug-versus-vehicle differential expression per contrast, with the release and request it answered.

  • get_perturbation_markers

    The genes a classifier found most important for one drug × cell line condition.

  • search_gene_sets

    Find gene sets (pathways) by name or collection, to pass to get_pathway_scores.

  • get_pathway_scores

    Differential VISION pathway scores per drug × cell line, with the release and request it answered.

  • get_phenotype_scores

    Survival, cell-cycle, heterogeneity and separability readouts for drugs × cell lines.

  • get_embedding_distance

    How far a drug moves a cell line's state from vehicle, in the release's embedding space.

  • run_sql

    Read-only SQL over the approved tables, guarded and cost-bounded; the exact SQL is returned.

  • get_guide

    How to analyse and interpret the data: controls, units, coverage traps.

  • health

    Liveness, the served release, and the response contract version.