startup-valuation

Biotech Pipeline Valuation

valuation_biotech

Read-onlyIdempotent

Value biotech drug pipelines with risk-adjusted methods: peak sales, decision-tree EV, or rNPV. Supply method-specific parameters to get results and assumptions.

Instructions

Risk-adjusted biotech valuation: peak sales, decision-tree expected value, and full pipeline rNPV across drugs. Method selects the model. Use for pharma/drug pipelines; for hardware or deep tech use valuation_hardware. Parameters apply per method: peak_sales needs patient_population + penetration + price; decision_tree needs probabilities + terminal_value; pipeline needs drugs + discount_rate. Not for hardware or deep tech — for that use valuation_hardware. Only method is required; all other parameters are method-dependent, so supply those the selected method names and omit the rest (defaults apply where defined). Rate and decimal inputs are fractions (0.10 = 10%); probability and weight lists are in [0,1] and sum to 1. Returns value, method, inputs, assumptions, chapter, formula_number and calculation steps; pure arithmetic — no I/O and no external calls — rounded to 2 decimals, with no auth or rate limits. An unknown method, or a missing method-required parameter, returns an error instead of a value.

Input Schema

TableJSON Schema

Name	Required	Description
`drugs`	No	Pipeline drugs: {name, peak_sales, probability, years_to_market, multiple(optional)}.
`price`	No	Price per unit / treatment, currency units.
`method`	Yes	Formula to apply. Options: peak_sales = Peak = population × penetration × price × compliance.; decision_tree = EV = Π pᵢ × terminal value.; pipeline = V = Σ(peak sales × multiple × P_success) / (1+r)^n.
`compliance`	No	Compliance / adherence rate as a decimal.
`penetration`	No	Market penetration as a decimal (0.10 = 10%).
`discount_rate`	No	Discount rate as a decimal (0.12 = 12%).
`probabilities`	No	Probability of each outcome or stage, each in [0,1]; the list must sum to 1 where it is exhaustive.
`terminal_value`	No	Expected exit / terminal value, currency units.
`patient_population`	No	Target patient population treated per year.

Output Schema

TableJSON Schema

Name	Required	Description
`error`	No	Error message when the call fails.
`steps`	No	Intermediate steps for traceability.
`value`	Yes	Computed valuation or metric.
`inputs`	No	Echo of the normalised inputs used.
`method`	No	Formula / method name that produced the result.
`chapter`	No	Source textbook chapter.
`assumptions`	No	Modelling assumptions applied.
`formula_number`	No	Source textbook formula number (e.g. '3.1').
`defaults_applied`	No	Optional parameters that were not supplied, so their documented defaults were used.

Schema Changelog

Changes observed during successful MCP inspections.

Changed1 schema field changedv2.1.1Oct 2, 2026

addedOutput schema / properties / defaults_applied

Added value: +{
+  "description": "Optional parameters that were not supplied, so their documented defaults were used.",
+  "items": {
+    "type": "string"
+  },
+  "type": "array"
+}

First observedv0.1.0Sep 30, 2026

TDQS

A4.7/5.0

Behavior5/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Annotations already declare readOnly/idempotent/no open world, and the description adds real behavioral context beyond them: pure arithmetic with no I/O or external calls, rounding to 2 decimals, no auth or rate limits, and explicit error behavior for an unknown method or a missing method-required parameter.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness3/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Front-loaded and information-dense, but the hard-tech exclusion is stated twice ('for hardware or deep tech use valuation_hardware' early, then 'Not for hardware or deep tech — for that use valuation_hardware' later), and the method-parameter mapping is buried in a long run-on middle sentence.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness5/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a 9-param, method-switching calculator with an output schema present, the description covers selection, per-method parameter requirements, edge-case errors, and numerical conventions. Nothing an agent needs to call it correctly is missing.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters4/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema description coverage is 100%, so the baseline is 3, but the description adds the conditional dependency structure the flat schema cannot express: peak_sales needs patient_population + penetration + price, decision_tree needs probabilities + terminal_value, pipeline needs drugs + discount_rate. It also clarifies fraction semantics (0.10 = 10%) and that unselected params should be omitted so defaults apply.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

Names a specific verb and resource ('risk-adjusted biotech valuation') and enumerates the three concrete models it produces (peak sales, decision tree, rNPV). It also explicitly separates itself from the sibling valuation_hardware, so an agent can route without opening a schema.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines5/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

States the domain ('pharma/drug pipelines'), the exclusion ('not for hardware or deep tech'), and names the alternative tool to use instead. The 'method selects the model' framing plus the required-only-'method' note tells the agent exactly how invocation works.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Deploy Server